images images listlengths 1 1 | data_source stringclasses 1
value | reward_model dict | extra_info dict | prompt listlengths 2 2 | ability stringclasses 1
value | env_name stringclasses 1
value |
|---|---|---|---|---|---|---|
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the hot dog is on the right side of the cup."
} | {
"answer": "Yes, the hot dog is on the right side of the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the hot dog on the right side of the cup\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"tab... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The handbag is to the left of the dog."
} | {
"answer": "The handbag is to the left of the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"ha... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the handbag to the left or right of the dog\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the handbag is on the left side of the bottle."
} | {
"answer": "Yes, the handbag is on the left side of the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_rel... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the handbag on the left side of the bottle\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices,... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the dog is not above the motorcycle."
} | {
"answer": "No, the dog is not above the motorcycle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the dog above the motorcycle\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</an... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the traffic light is above the person."
} | {
"answer": "Yes, the traffic light is above the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the traffic light above the person\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the car is not on the left side of the person."
} | {
"answer": "No, the car is not on the left side of the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_rela... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the car on the left side of the person\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The car is below the traffic light."
} | {
"answer": "The car is below the traffic light.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_t... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the car above or below the traffic light\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"h... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the car is above the chair."
} | {
"answer": "Yes, the car is above the chair.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the car above the chair\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"tire\": 2, \... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The wine glass is above the knife."
} | {
"answer": "The wine glass is above the knife.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the wine glass above or below the knife\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The backpack is above the dog."
} | {
"answer": "The backpack is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_target... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the backpack above or below the dog\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The person is above the dog."
} | {
"answer": "The person is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the person above or below the dog\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot dog\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dog is to the left of the bottle."
} | {
"answer": "The dog is to the left of the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the dog to the left or right of the bottle\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices,... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the bottle is not to the left of the person."
} | {
"answer": "No, the bottle is not to the left of the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relati... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bottle to the left of the person\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dog is below the book."
} | {
"answer": "The dog is below the book.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets": ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the dog above or below the book\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}<... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the book is not below the bottle."
} | {
"answer": "No, the book is not below the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the book below the bottle\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</ans... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the dog is not above the traffic light."
} | {
"answer": "No, the dog is not above the traffic light.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the dog above the traffic light\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"tire... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the truck is above the dog."
} | {
"answer": "Yes, the truck is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the truck above the dog\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <answer>{\"fl... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the apple is not below the oven."
} | {
"answer": "No, the apple is not below the oven.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the apple below the oven\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</answ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bottle is above the apple."
} | {
"answer": "The bottle is above the apple.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_target... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the bottle above or below the apple\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dining table is below the cup."
} | {
"answer": "The dining table is below the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the dining table above or below the cup\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dog is to the left of the cell phone."
} | {
"answer": "The dog is to the left of the cell phone.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the dog to the left or right of the cell phone\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answe... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The cell phone is above the bird."
} | {
"answer": "The cell phone is above the bird.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_tar... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the cell phone above or below the bird\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the bird is below the dog."
} | {
"answer": "Yes, the bird is below the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targe... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bird below the dog\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</answer... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The truck is above the fire hydrant."
} | {
"answer": "The truck is above the fire hydrant.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the truck above or below the fire hydrant\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"ta... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the truck is above the stop sign."
} | {
"answer": "Yes, the truck is above the stop sign.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"ha... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the truck above the stop sign\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot dog\": 1... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the fire hydrant is not to the left of the stop sign."
} | {
"answer": "No, the fire hydrant is not to the left of the stop sign.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spati... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the fire hydrant to the left of the stop sign\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The sink is on the right side of the bottle."
} | {
"answer": "The sink is on the right side of the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink on the left or right side of the bottle\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bottle is below the dog."
} | {
"answer": "The bottle is below the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the bottle above or below the dog\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot dog\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the bottle is not above the backpack."
} | {
"answer": "No, the bottle is not above the backpack.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bottle above the backpack\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}<... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dog is to the right of the backpack."
} | {
"answer": "The dog is to the right of the backpack.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the dog to the left or right of the backpack\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indice... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the handbag is on the right side of the dog."
} | {
"answer": "Yes, the handbag is on the right side of the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the handbag on the right side of the dog\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, l... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The stop sign is above the person."
} | {
"answer": "The stop sign is above the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the stop sign above or below the person\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, li... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the clock is not on the left side of the bottle."
} | {
"answer": "No, the clock is not on the left side of the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_re... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the clock on the left side of the bottle\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"h... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The dog is to the left of the person."
} | {
"answer": "The dog is to the left of the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the dog to the left or right of the person\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the oven is not above the cup."
} | {
"answer": "No, the oven is not above the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the oven above the cup\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <answer>{\"flo... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bowl is on the left side of the laptop."
} | {
"answer": "The bowl is on the left side of the laptop.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bowl on the left or right side of the laptop\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image in... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The cup is on the left side of the microwave."
} | {
"answer": "The cup is on the left side of the microwave.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the cup on the left or right side of the microwave\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the cup is above the dog."
} | {
"answer": "Yes, the cup is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_target... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the cup above the dog\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"tire\": 2, \"r... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the surfboard is not to the right of the dog."
} | {
"answer": "No, the surfboard is not to the right of the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the surfboard to the right of the dog\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The clock is above the handbag."
} | {
"answer": "The clock is above the handbag.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targe... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the clock above or below the handbag\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot d... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the dog is not above the backpack."
} | {
"answer": "No, the dog is not above the backpack.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"ha... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the dog above the backpack\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</answ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The person is above the dog."
} | {
"answer": "The person is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the person above or below the dog\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot dog\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the motorcycle is to the left of the chair."
} | {
"answer": "Yes, the motorcycle is to the left of the chair.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relati... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the motorcycle to the left of the chair\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, li... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the dog is to the left of the person."
} | {
"answer": "Yes, the dog is to the left of the person.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the dog to the left of the person\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"ti... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the handbag is above the dog."
} | {
"answer": "Yes, the handbag is above the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the handbag above the dog\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <answer>{\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the dog is on the right side of the car."
} | {
"answer": "Yes, the dog is on the right side of the car.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the dog on the right side of the car\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the bowl is to the left of the dog."
} | {
"answer": "Yes, the bowl is to the left of the dog.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the bowl to the left of the dog\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"tire... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The handbag is below the backpack."
} | {
"answer": "The handbag is below the backpack.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the handbag above or below the backpack\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"t... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the backpack is on the left side of the car."
} | {
"answer": "Yes, the backpack is on the left side of the car.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the backpack on the left side of the car\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, l... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the handbag is to the left of the car."
} | {
"answer": "Yes, the handbag is to the left of the car.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the handbag to the left of the car\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the microwave is above the bottle."
} | {
"answer": "Yes, the microwave is above the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"h... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the microwave above the bottle\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bottle is below the clock."
} | {
"answer": "The bottle is below the clock.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_target... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the bottle above or below the clock\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\":... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The sink is above the bottle."
} | {
"answer": "The sink is above the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink above or below the bottle\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The vase is above the toilet."
} | {
"answer": "The vase is above the toilet.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the vase above or below the toilet\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the sink is not above the clock."
} | {
"answer": "No, the sink is not above the clock.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink above the clock\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <answer>{\"f... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the cup is not above the sink."
} | {
"answer": "No, the cup is not above the sink.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the cup above the sink\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</answer>\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the microwave is on the right side of the bottle."
} | {
"answer": "Yes, the microwave is on the right side of the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the microwave on the right side of the bottle\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the oven is not below the cup."
} | {
"answer": "No, the oven is not below the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the oven below the cup\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"tire\": 2, \"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The oven is below the cup."
} | {
"answer": "The oven is below the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets": ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the oven above or below the cup\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}<... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The cup is above the orange."
} | {
"answer": "The cup is above the orange.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the cup above or below the orange\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"ti... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the oven is on the right side of the orange."
} | {
"answer": "Yes, the oven is on the right side of the orange.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the oven on the right side of the orange\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"tab... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The chair is below the sink."
} | {
"answer": "The chair is below the sink.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the chair above or below the sink\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <an... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the sink is not to the right of the toaster."
} | {
"answer": "No, the sink is not to the right of the toaster.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relati... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the sink to the right of the toaster\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the bowl is on the right side of the wine glass."
} | {
"answer": "Yes, the bowl is on the right side of the wine glass.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_r... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the bowl on the right side of the wine glass\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The wine glass is below the bottle."
} | {
"answer": "The wine glass is below the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_t... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the wine glass above or below the bottle\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"tab... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the bottle is not above the scissors."
} | {
"answer": "No, the bottle is not above the scissors.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bottle above the scissors\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}<... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The oven is below the wine glass."
} | {
"answer": "The oven is below the wine glass.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_tar... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the oven above or below the wine glass\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tr... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the potted plant is not below the wine glass."
} | {
"answer": "No, the potted plant is not below the wine glass.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the potted plant below the wine glass\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"hot ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the oven is below the potted plant."
} | {
"answer": "Yes, the oven is below the potted plant.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the oven below the potted plant\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <answ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the potted plant is below the orange."
} | {
"answer": "Yes, the potted plant is below the orange.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the potted plant below the orange\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, like <an... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The sink is below the potted plant."
} | {
"answer": "The sink is below the potted plant.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_t... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the sink above or below the potted plant\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the sink is not below the vase."
} | {
"answer": "No, the sink is not below the vase.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_t... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink below the vase\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</answe... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the sink is below the microwave."
} | {
"answer": "Yes, the sink is below the microwave.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the sink below the microwave\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</an... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The oven is to the right of the microwave."
} | {
"answer": "The oven is to the right of the microwave.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the oven to the left or right of the microwave\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1,... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the clock is not below the oven."
} | {
"answer": "No, the clock is not below the oven.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the clock below the oven\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</answ... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "Yes, the sink is below the clock."
} | {
"answer": "Yes, the sink is below the clock.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_tar... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the sink below the clock\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}</answer... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bowl is on the left side of the sandwich."
} | {
"answer": "The bowl is on the left side of the sandwich.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the bowl on the left or right side of the sandwich\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <a... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The bowl is on the right side of the sink."
} | {
"answer": "The bowl is on the right side of the sink.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bowl on the left or right side of the sink\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\":... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The cup is above the sink."
} | {
"answer": "The cup is above the sink.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets": ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThe sentence is \"Is the cup above or below the sink\". First locate all relevant instances, then use zoom-in tool on each instance. Answer with at most the top 3 instances and their zoom indices, like <answer>{\"person\": 1, \"table\": 2}<... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The microwave is above the bottle."
} | {
"answer": "The microwave is above the bottle.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_ta... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the microwave above or below the bottle\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, li... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the bottle is not above the toaster."
} | {
"answer": "No, the bottle is not above the toaster.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the bottle above the toaster\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\": 2}</... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The truck is to the left of the cup."
} | {
"answer": "The truck is to the left of the cup.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the truck to the left or right of the cup\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <answer>{\"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the cup is not on the left side of the car."
} | {
"answer": "No, the cup is not on the left side of the car.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relatio... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the cup on the left side of the car\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tree\... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the cell phone is not below the backpack."
} | {
"answer": "No, the cell phone is not below the backpack.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation"... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nFor the sentence \"Is the cell phone below the backpack\". First find all instances that are related, then zoom-in on those instances to see details. List the top 3 instances with their zoomed image indices, like <answer>{\"wheel\": 1, \"ti... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the dining table is not to the left of the sink."
} | {
"answer": "No, the dining table is not to the left of the sink.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_re... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the dining table to the left of the sink\". First find all related instances, then use zoom-in tool on those instances to see them clearly. Answer with at most the top 3 instances and their zoomed image indices, l... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the microwave is not on the right side of the dining table."
} | {
"answer": "No, the microwave is not on the right side of the dining table.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": ... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nGiven the sentence \"Is the microwave on the right side of the dining table\". First identify all related instances, then zoom-in on those instances to examine them closely. Report the top 3 instances with their zoomed image indices, like <... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "No, the sink is not to the left of the microwave."
} | {
"answer": "No, the sink is not to the left of the microwave.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 2,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relat... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink to the left of the microwave\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"tre... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 | |
curriculum_1_sealvqa_spatial | {
"ground_truth": "The sink is below the potted plant."
} | {
"answer": "The sink is below the potted plant.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes use absolute pixel coordinates in [x, y, width, height] format",
"conversation_turns": 4,
"curriculum_stage": 1,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_t... | [
{
"content": "",
"role": "system"
},
{
"content": "<image>\nThis is the sentence: \"Is the sink above or below the potted plant\". First identify the relevant instances, then apply zoom-in tool on each instance. Report at most the top 3 instances and their zoom indices, like <answer>{\"bird\": 1, \"... | curriculum_1_sealvqa_spatial_spatial | visual_toolbox_v2 |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 9