Dataset Viewer
images
images list | data_source
string | reward_model
dict | extra_info
dict | prompt
list | ability
string | env_name
string | bbox_format
string | source_dataset
string | question
string | image
string | answer
string | target_instances
list | conversations
list | search
int64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
sealvqa_spatial
|
{
"ground_truth": "No, the down more head is not on the right side of the horse with 2 straps on face."
}
|
{
"answer": "No, the down more head is not on the right side of the horse with 2 straps on face.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes are in absolute pixel coordinates with format [x, y, width, height], where (x, y) is the top-left corner, width is horizontal extent, and height is vertical extent.",
"conversation_turns": 2,
"data_split": "train",
"dataset_type": "spatial_relation",
"has_targets": true,
"image_path": "coco2014/train2014/COCO_train2014_000000127622.jpg",
"index": "sealvqa_spatial_41905",
"original_conversations": [
{
"from": "human",
"value": "<image>\nAdditional visual information to focus on: <object>; <object>.\nIs the down more head on the right side of the horse with 2 straps on face?"
},
{
"from": "gpt",
"value": "No, the down more head is not on the right side of the horse with 2 straps on face."
}
],
"question": "the object; the object.\nIs the down more head on the right side of the horse with 2 straps on face?",
"question_id": "41905",
"search": 1,
"target_instances": [
{
"bbox": [
227.65,
209.3,
144.8,
345.37
],
"instance_id": "",
"name": "horse with 2 straps on face"
},
{
"bbox": [
67.15,
234.54,
127.45,
357.69
],
"instance_id": "",
"name": "down more head"
}
]
}
|
[
{
"content": "You are a helpful assistant.\n\n# Tools\nYou may call one or more functions to assist with the user query.\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\":\"function\",\"function\":{\"name\":\"image_zoom_in_tool\",\"description\":\"Zoom in on a specific region of an image by cropping it based on a bounding box (bbox) and an optional object label.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"bbox_2d\":{\"type\":\"array\",\"items\":{\"type\":\"number\"},\"minItems\":4,\"maxItems\":4,\"description\":\"The bounding box of the region to zoom in, as [x1, y1, x2, y2], where (x1, y1) is the top-left corner and (x2, y2) is the bottom-right corner.\"},\"label\":{\"type\":\"string\",\"description\":\"The name or label of the object in the specified bounding box (optional).\"}},\"required\":[\"bbox\"]}}}\n</tools>\n\n# How to call a tool\nReturn a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call>\n\n**Example**: \n<tool_call> \n{\"name\": \"image_zoom_in_tool\", \"arguments\": {\"bbox_2d\": [10, 20, 100, 200], \"label\": \"the apple on the desk\"}} \n</tool_call>",
"role": "system"
},
{
"content": "<image>\nthe object; the object.\nIs the down more head on the right side of the horse with 2 straps on face?\nThink first, call **image_zoom_in_tool** if needed, then answer. Format strictly as: <think>...</think> <tool_call>...</tool_call> (if tools needed) <answer>...</answer> ",
"role": "user"
}
] |
sealvqa_spatial_spatial
|
visual_toolbox_v2
|
absolute_xywh
|
sealvqa_spatial
| null | null | null | null | null | null |
|
sealvqa_gqa
|
{
"ground_truth": "The ball is to the right of the man that is wearing socks."
}
|
{
"answer": "The ball is to the right of the man that is wearing socks.",
"bbox_format": "absolute_xywh",
"bbox_format_description": "Bounding boxes are in absolute pixel coordinates with format [x, y, width, height], where (x, y) is the top-left corner, width is horizontal extent, and height is vertical extent.",
"conversation_turns": 2,
"data_split": "train",
"dataset_type": "gqa",
"has_targets": true,
"image_path": "gqa/images/2361897.jpg",
"index": "sealvqa_gqa_41905",
"original_conversations": [
{
"from": "human",
"value": "<image>\nAdditional visual information to focus on: <object>; <object>.\nIs the ball to the left or to the right of the man that is wearing socks?"
},
{
"from": "gpt",
"value": "The ball is to the right of the man that is wearing socks."
}
],
"question": "Is the ball to the left or to the right of the man that is wearing socks?",
"question_id": "41905",
"search": 1,
"target_instances": [
{
"bbox": [
299,
166,
27,
26
],
"instance_id": "2224233",
"name": "ball"
},
{
"bbox": [
122,
188,
11,
17
],
"instance_id": "1793964",
"name": "socks"
}
]
}
|
[
{
"content": "You are a helpful assistant.\n\n# Tools\nYou may call one or more functions to assist with the user query.\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>\n{\"type\":\"function\",\"function\":{\"name\":\"image_zoom_in_tool\",\"description\":\"Zoom in on a specific region of an image by cropping it based on a bounding box (bbox) and an optional object label.\",\"parameters\":{\"type\":\"object\",\"properties\":{\"bbox_2d\":{\"type\":\"array\",\"items\":{\"type\":\"number\"},\"minItems\":4,\"maxItems\":4,\"description\":\"The bounding box of the region to zoom in, as [x1, y1, x2, y2], where (x1, y1) is the top-left corner and (x2, y2) is the bottom-right corner.\"},\"label\":{\"type\":\"string\",\"description\":\"The name or label of the object in the specified bounding box (optional).\"}},\"required\":[\"bbox\"]}}}\n</tools>\n\n# How to call a tool\nReturn a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call>\n\n**Example**: \n<tool_call> \n{\"name\": \"image_zoom_in_tool\", \"arguments\": {\"bbox_2d\": [10, 20, 100, 200], \"label\": \"the apple on the desk\"}} \n</tool_call>",
"role": "system"
},
{
"content": "<image>\nIs the ball to the left or to the right of the man that is wearing socks?\nThink first, call **image_zoom_in_tool** if needed, then answer. Format strictly as: <think>...</think> <tool_call>...</tool_call> (if tools needed) <answer>...</answer> ",
"role": "user"
}
] |
sealvqa_gqa_spatial
|
visual_toolbox_v2
|
absolute_xywh
|
sealvqa_gqa
|
Is the ball to the left or to the right of the man that is wearing socks?
|
gqa/images/2361897.jpg
|
The ball is to the right of the man that is wearing socks.
|
[
{
"bbox": [
299,
166,
27,
26
],
"instance_id": "2224233",
"name": "ball"
},
{
"bbox": [
122,
188,
11,
17
],
"instance_id": "1793964",
"name": "socks"
}
] |
[
{
"from": "human",
"value": "<image>\nAdditional visual information to focus on: <object>; <object>.\nIs the ball to the left or to the right of the man that is wearing socks?"
},
{
"from": "gpt",
"value": "The ball is to the right of the man that is wearing socks."
}
] | 1 |
README.md exists but content is empty.
- Downloads last month
- 92