| { |
| "source": "5_template1", |
| "questions": [ |
| { |
| "subject": "sculpture", |
| "part": "material", |
| "level": "global_attribute", |
| "relevance": "text_and_image", |
| "reward": 5, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to the material of the sculpture?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| }, |
| { |
| "subject": "sculpture", |
| "part": "color", |
| "level": "global_attribute", |
| "relevance": "text_and_image", |
| "reward": 5, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to the color of the sculpture?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| }, |
| { |
| "subject": "sculpture", |
| "part": "stagehand", |
| "level": "part", |
| "relevance": "text_and_image", |
| "reward": 4, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to stagehand?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| }, |
| { |
| "subject": "sculpture", |
| "part": "water bottle", |
| "level": "part", |
| "relevance": "text_and_image", |
| "reward": 4, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to water bottle?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| }, |
| { |
| "subject": "sculpture", |
| "part": "gloves", |
| "level": "part", |
| "relevance": "text_and_image", |
| "reward": 4, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to gloves?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| }, |
| { |
| "subject": "sculpture", |
| "part": "puddles", |
| "level": "part", |
| "relevance": "text_and_image", |
| "reward": 4, |
| "question": { |
| "question_type": "multimodal_alignment", |
| "input": "reference_text + reference_image + generated_image", |
| "question": "Does the generated image correctly represent the expected result of applying the reference text to the reference image with respect to puddles?", |
| "answer_space": [ |
| "Yes", |
| "No" |
| ] |
| } |
| } |
| ] |
| } |
