data_source stringclasses 1
value | prompt listlengths 1 1 | images images listlengths 1 4 | ability stringclasses 1
value | reward_model dict | extra_info dict |
|---|---|---|---|---|---|
hiyouga/geometry3k | [
{
"content": "<image>\nAs the boy in white clothes in the photo, in which direction are you facing?\nOptions:\nA. left\nB. back-left\nC. front-right\nD. front-left\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_647_2",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in red. The camera or lighting is changed for the second image, which shows several red-circled points labeled 'A, B, C, D'. Which point is the match?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, ... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_78fab9be-f693-4a16-8cd2-7a6b4c33d2b1",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nWhich hand does this man hold the umbrella with?\nOptions:\nA. left\nB. right\nC. can not determine\nD. can not determine\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "omnispatial_663_1",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe left image shows the original cube stack made of equal-sized small cubes. Which of the options on the right cannot be obtained by rotating the original cube stack? Please answer from options A, B, C, or D.\nOptions:\nA. A\nB. B\nC. C\nD. All three other options are incorrect\nAnswer w... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "omnispatial_682_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is circled in gold in the first image. The camera or lighting is changed, and the second image displays several gold-circled points labeled with '1, 2, 3, 4'. Which corresponds to the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>.... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_22ca7060-ec64-43ea-8c33-0b044a2fcb29",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThese two images (image 1 and 2) show the same scene from two different viewpoints. The image 2 was taken after turning the camera 90 degrees to the left (counterclockwise) from the position of image 1. Based on these two images: If I am standing at the same spot and facing the s... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_rotation_group031_qsecond_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in pink. The second image, taken after altering the camera or lighting, presents several pink-circled points labeled '1, 2, 3, 4'. Which corresponds to the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_13205f00-409c-4f46-b7cc-b40a9abcbd92",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nFrom the perspective of the man in blue clothes, in which direction is the man in white clothes?\nOptions:\nA. right\nB. back-right\nC. left\nD. back\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_884_1",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image shows a gold-circled point. The camera or lighting is changed, and the second image contains multiple gold-circled points, each with a 'A, B, C, D' label. Which point is the same?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</th... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_64888882-a91c-496f-8a8e-236853fcd104",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in cyan. The camera or lighting is changed for the second image, which shows several cyan-circled points labeled '1, 2, 3, 4'. Which point is the match?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_50a08c84-8842-4ce5-b5cc-acbb3c41688a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image contains a point circled in brown. After changing the camera or lighting, the second image shows multiple brown-circled points labeled '1, 2, 3, 4'. Select the matching point.\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_21a3b921-17a0-4147-bbde-a55c0e2e871f",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a red circled point in image one. After changing the camera angle or lighting, image two is captured. Several points are circled in red and labeled with 'A, B, C, D'. Which point corresponds to the original?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by st... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_f92fd950-6e7b-4135-8b25-2c49226bdb27",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe original paper is a 3×3 grid paper. The images in the top row show the results of 2 consecutive folding operations on the grid paper. Folding operations include folding along horizontal, vertical, or 45-degree direction. The rightmost image in the top row shows the result after punchi... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "omnispatial_503_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a blue circle in the first image. After changing the camera or lighting, the second image shows multiple blue-circled points labeled '1, 2, 3, 4'. Which matches?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>, ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_f3c73759-a564-4e0f-85ac-23a467822432",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the small glass jar from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I am standing at the same spot and facin... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_among_group395_q2_2_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image shows a green-circled point. The camera or lighting is changed, and the second image contains multiple green-circled points, each with a 'A, B, C, D' label. Which point is the same?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_43e7186b-2c91-4980-8994-cd3ab025b51a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is marked with a pink circle in the first image. The second image, taken after changing the camera or lighting, shows several pink-circled points labeled with 'A, B, C, D'. Select the matching point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <t... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_5bba87b9-f3c9-4b7c-b897-389b31aa265a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nWhen positioned at window facing desk, where can you find paper?\nOptions:\nA. back-right\nB. right\nC. left\nD. front\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "omnispatial_230_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a point circled in brown in the first image. After changing the camera or lighting, the second image shows several brown-circled points labeled 'A, B, C, D'. Which one corresponds?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, t... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_751606b9-5f4e-4af0-b663-772327ca8bf0",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nTaking the camera's viewpoint as the front, which way is the man in white trousers facing in the image?\nOptions:\nA. left\nB. right\nC. back-left\nD. front\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"r... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "omnispatial_1065_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe original paper is a 3×3 grid paper. The images in the top row show the results of 2 consecutive folding operations on the grid paper. Folding operations include folding along horizontal, vertical, or 45-degree direction. The rightmost image in the top row shows the result after punchi... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "omnispatial_505_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a brown circle in the first image. After changing the camera or lighting, the second image shows multiple brown-circled points labeled 'A, B, C, D'. Which matches?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_47f9a2ca-6ed5-462f-9d71-e523d68121c0",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a blue circle in image one. After adjusting the camera or lighting, image two presents multiple blue-circled points labeled 'A, B, C, D'. Which matches the original point?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_03074343-14df-43f9-8d7a-61de14f8ba4e",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is circled in orange in the first image. After adjusting the camera or lighting, the second image presents several orange-circled points labeled '1, 2, 3, 4'. Which is the matching point?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</th... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_457f3647-e31c-48bd-b748-1a7ab279ae06",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is marked with a blue circle in the first image. The second image, taken after changing the camera or lighting, shows several blue-circled points labeled with 'A, B, C, D'. Select the matching point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <t... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_11a4b066-2aaf-472d-aa2c-577325b1e493",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn image one, a orange point is circled. The camera position or lighting is changed before taking image two, which contains multiple orange-circled points labeled with '1, 2, 3, 4'. Which point matches the first?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_598a7904-3f5f-4ac6-b341-114054263968",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nBased on these two views showing the same scene: in which direction did I move from the first view to the second view? A. Diagonally forward and right B. Directly left C. Diagonally forward and left D. Directly right\nAnswer with one option letter.\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_among_group029_q1_1_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe top row shows the isometric view (left) and the top view (right) of a 3D model. Which image in the bottom row is the front view of the model? Please answer from options A, B, C, or D.\nOptions:\nA. A\nB. B\nC. C\nD. All three other options are incorrect\nAnswer with one option letter.... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_812_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nI am entering a 30 zone:\nOptions:\nA. Yes\nB. No\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "omnispatial_1333_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/spatial_interaction",
"tool_label": "mixed",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (front, left, and right): If I am standing at the same spot and facing the same direction as shown in image 2, then I turn right and go straight, will I get closer to the lap... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group075_q3_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nThese four images (image 1, 2, 3, and 4) show the same scene from different viewpoints. The camera basically stood in one spot and turned 90 degrees each time to get all sides of the area. Based on these four images: If I am standing at the same spot and facing ... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "mindcube_rotation_group011_q3_4",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nBased on these two views showing the same scene: in which direction did I move from the first view to the second view? A. Diagonally forward and left B. Directly left C. Directly right D. Diagonally forward and right\nAnswer with one option letter.\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_among_group461_q1_1_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image shows a point circled in green. After adjusting the camera or lighting, the second image presents several green-circled points labeled 'A, B, C, D'. Which matches the original?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_a7b2c4f5-6604-4dd7-a7af-beacce83fc30",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (front, left, and right): If I am standing at the same spot and facing the same direction as shown in image 2, then I turn right and go straight, will I get closer to the squ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group021_q3_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a point circled in brown in the first image. After changing the camera or lighting, the second image shows several brown-circled points labeled '1, 2, 3, 4'. Which one corresponds?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>, t... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_cfdf9596-be33-4884-9d8f-438f29a89fa3",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in cyan. The camera or lighting is changed for the second image, which shows several cyan-circled points labeled '1, 2, 3, 4'. Which point is the match?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_f18585fb-ce34-4dcd-8fa1-f9fe3c3fe59c",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in green. The second image, taken after altering the camera or lighting, presents several green-circled points labeled 'A, B, C, D'. Which corresponds to the original?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <th... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_e928521e-486d-4ae2-bb9e-e41b532c060d",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (back, left, and right): If I am standing at the same spot and facing the same direction as shown in image 3, then I turn right and go straight, what would happen to my relat... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group086_q3_10",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image features a point circled in orange. After adjusting the camera or lighting, the second image contains several orange-circled points labeled '1, 2, 3, 4'. Which point is the same?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</thi... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_2f760533-4eda-42da-b60e-315886e3ad5e",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn image one, a blue point is circled. The camera position or lighting is changed before taking image two, which contains multiple blue-circled points labeled with '1, 2, 3, 4'. Which point matches the first?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_67b88afc-5c9c-4c72-a87e-73c4225bd736",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the initial image, a point is circled in orange. The second image is taken after changing the camera or lighting, and it has several orange-circled points labeled with '1, 2, 3, 4'. Which corresponds to the first?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by s... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_d9717b69-d661-41c2-b428-f049ada59386",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA orange circled point is present in the first image. The second image, taken after altering the camera or lighting, features several orange-circled points with 'A, B, C, D' labels. Which is the corresponding point?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by st... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_c7e2234d-47d9-4b81-9297-970bd913ff0a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in blue. The second image, taken after altering the camera or lighting, presents several blue-circled points labeled 'A, B, C, D'. Which corresponds to the original?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_f1f4ea78-4422-4e3f-b883-d24bbf868bfc",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a blue circle in the first image. After changing the camera or lighting, the second image shows multiple blue-circled points labeled 'A, B, C, D'. Which matches?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, ... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_d2d5b232-2d91-46cf-9a00-05936a90dcf6",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (front, left, and right): From the viewpoint presented in image 3, is there a printer behind the wooden cabinet? A. Yes B. No.\nAnswer with one option letter.\n\nReason step ... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group093_q2_4",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nFrom the perspective of this woman, where is the turkey positioned?\nOptions:\nA. back-left\nB. right\nC. front\nD. left\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_1143_1",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nGiven three views, how many cubes could be needed to satisfy the constraints shown in the images? Please answer from options A, B, C, or D.\nOptions:\nA. 12\nB. 7\nC. 11\nD. 4\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</fina... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "omnispatial_1114_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is marked with a cyan circle in the first image. The second image, taken after changing the camera or lighting, shows several cyan-circled points labeled with 'A, B, C, D'. Select the matching point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <t... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_840299d0-a3f9-444b-a845-4779e85452c1",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the toy train from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I was positioned where the toy train is and fa... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "mindcube_among_group740_gen_4_2",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (front, left, and right): If I am standing at the same spot and facing the same direction as shown in image 3, then I turn left and go straight, what would happen to my relat... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group053_q3_2",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nFrom the perspective of the man in red clothes, where is the white car?\nOptions:\nA. back-right\nB. front-left\nC. front\nD. back-left\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "omnispatial_1081_1",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the red jar from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I am standing at the same spot and facing the sa... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_among_group485_q3_2_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in cyan. The camera position or lighting is altered, and a second image is captured. Multiple cyan-circled points with '1, 2, 3, 4' labels appear in the second image. Identify the corresponding point.\nA point-1\nB point-2\nC point-3\nD poin... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_2a85cec7-3dba-4454-b728-5f2b18a78255",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a brown circled point in image one. After changing the camera angle or lighting, image two is captured. Several points are circled in brown and labeled with '1, 2, 3, 4'. Which point corresponds to the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step b... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_51805a29-cddc-4baa-816f-65cf655016e5",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the initial image, a point is circled in orange. The second image is taken after changing the camera or lighting, and it has several orange-circled points labeled with '1, 2, 3, 4'. Which corresponds to the first?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by s... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_f683b233-9a32-42d5-b096-1491bff9a7c8",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a pink circled point in image one. After changing the camera angle or lighting, image two is captured. Several points are circled in pink and labeled with '1, 2, 3, 4'. Which point corresponds to the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by ... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_72c7258c-103d-4d41-92f7-fd20a8a4ac1a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn image one, a point is circled in green. The camera or lighting is changed, and image two contains multiple green-circled points labeled 'A, B, C, D'. Identify the matching point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, then out... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_9635765f-d065-4c32-adc7-ed9ebe0351b1",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a point circled in cyan in the first image. After changing the camera or lighting, the second image shows several cyan-circled points labeled 'A, B, C, D'. Which one corresponds?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, the... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_43548408-cb52-4d0e-8f63-bf161866817b",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nThese four images (image 1, 2, 3, and 4) show the same scene from different viewpoints. The camera basically stood in one spot and turned 90 degrees each time to get all sides of the area. Based on these four images: If I am standing at the same spot and facing ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_rotation_group033_q1_8",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nPicture yourself as the zebra; which way are you looking in the scene?\nOptions:\nA. back\nB. front\nC. back-left\nD. right\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "omnispatial_826_1",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image features a green-circled point. After altering the camera or lighting, the second image displays several green-circled points with '1, 2, 3, 4' labels. Which one matches the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_f91b9a9e-21d6-4b37-bb22-fa07649c6b65",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the blue bottle from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I was positioned where the light-colored sof... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_among_group516_gen_6_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the gray sneaker from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: From the viewpoint presented in image 4, what ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_among_group685_q3_5_2",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThese two images (image 1 and 2) show the same scene from two different viewpoints. Image 2 was taken after turning the camera 90 degrees to the right (clockwise) from the position of image 1. Based on these two images: If I am standing at the same spot and facing the same direct... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_rotation_group038_qsecond_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe top row shows the combined shape viewed from two different angles. The shape consists of a circular frustum, a square frustum, and a triangular prism from top to bottom. Which of the following images cannot be a cross-section of the shape? Please answer from options A, B, C, or D.\nOp... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "omnispatial_935_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point circled in orange is shown in the first image. After changing the camera or lighting, the second image has several orange-circled points labeled 'A, B, C, D'. Which matches the first?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_384b470a-951a-439a-9f6f-2a64e0b1b026",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in green. The camera position or lighting is altered, and a second image is captured. Multiple green-circled points with '1, 2, 3, 4' labels appear in the second image. Identify the corresponding point.\nA point-1\nB point-2\nC point-3\nD po... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_b69cfd84-c590-4174-b83d-773c1c7d2adb",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the yellow can from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: From the viewpoint presented in image 1, what is... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "mindcube_among_group494_q0_5_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point circled in pink appears in the first image. The second image, taken with a different camera position or lighting, shows several pink-circled points labeled 'A, B, C, D'. Identify the corresponding point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step i... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_54b89479-a31a-42d6-8653-91cf78cb6b99",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the yellow ball from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I am standing at the same spot and facing th... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_among_group142_q0_2_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThese two images (image 1 and 2) show the same scene from two different viewpoints. Image 2 was taken after turning the camera 90 degrees to the right (clockwise) from the position of image 1. Based on these two images: If I am standing at the same spot and facing the same direct... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_rotation_group040_qfirst_5",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is circled in green in the first image. After adjusting the camera or lighting, the second image presents several green-circled points labeled 'A, B, C, D'. Which is the matching point?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</thin... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_1ca24bf0-b8b5-4e7f-9913-e7d9b21383c1",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nThese three images (image 1, 2, and 3) show the same scene from three different viewpoints. The image 2 was taken after turning the camera 90 degrees to the right (clockwise) from the position of image 1. For image 3, the camera was turned another 90 degrees right, so it... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_rotation_group009_q3_7",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in cyan. The second image, taken after altering the camera or lighting, presents several cyan-circled points labeled '1, 2, 3, 4'. Which corresponds to the original?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_9ce5bb38-4850-4be0-b058-cb3f2920a54c",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image contains a cyan-circled point. After adjusting the camera or lighting, the second image displays several cyan-circled points with '1, 2, 3, 4' labels. Which is the corresponding point?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>..... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_74edbcb5-5519-47a9-9acb-89b31f345d25",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in green. The camera or lighting is changed for the second image, which shows several green-circled points labeled '1, 2, 3, 4'. Which point is the match?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</thin... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_e24455d1-40a1-4a32-8751-1ca9cda0762a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA pink point is circled on the first image. After changing the camera position or lighting, a second image is taken. Several pink-circled points are visible in the second image, each labeled with 'A, B, C, D'. Which one matches the original point?\nA point-A\nB point-B\nC point-C... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_1b90a37e-2d76-4e74-a49b-6b5d0e9a0111",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nBased on these two views showing the same scene: in which direction did I move from the first view to the second view? A. Directly right B. Diagonally forward and left C. Directly left D. Diagonally forward and right\nAnswer with one option letter.\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group001_q1_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is circled in blue in the first image. After adjusting the camera or lighting, the second image presents several blue-circled points labeled '1, 2, 3, 4'. Which is the matching point?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_91d386a3-1f27-49ea-8ea9-e09ad052020a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a cyan circle in the first image. After changing the camera or lighting, the second image shows multiple cyan-circled points labeled '1, 2, 3, 4'. Which matches?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>, ... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_f6740f84-804a-40ac-b76a-29f429836281",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a point circled in cyan in the first image. After changing the camera or lighting, the second image shows several cyan-circled points labeled 'A, B, C, D'. Which one corresponds?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, the... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_bb9812a5-e0ba-415f-97c5-f431b08413d5",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nThe original paper is a 5×5 grid paper. The images in the top row show the results of 3 consecutive folding operations on the grid paper. Folding operations include folding along horizontal, vertical, or 45-degree direction. The rightmost image in the top row shows the result after punchi... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_561_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/complex_logic_no_tool",
"tool_label": "no_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\nStanding at curtain, gazing at sofa, where should bed be?\nOptions:\nA. left\nB. back-left\nC. front-right\nD. front-left\nAnswer with one option letter.\n\nReason step by step in <think>...</think>, then output <final_answer>...</final_answer>.",
"role": "user"
}
] | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "omnispatial_90_0",
"original_data_source": "omnispatial",
"source_subset": "omnispatial/perspective_taking",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nBased on these two views showing the same scene: in which direction did I move from the first view to the second view? A. Directly right B. Directly left C. Diagonally forward and left D. Diagonally forward and right\nAnswer with one option letter.\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group036_q1_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA gold point is circled in the first image. The second image, taken after adjusting the camera or lighting, has several gold-circled points labeled '1, 2, 3, 4'. Identify the corresponding point.\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</th... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_66c3cbc6-4070-4016-aff0-94d0dba3c99a",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA green point is circled in image one. The camera or lighting is changed before taking image two, which displays several green-circled points labeled '1, 2, 3, 4'. Which matches the original point?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_148e08af-2810-4129-b663-e1bae4262f01",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\n<image>\nBased on these four images (image 1, 2, 3, and 4) showing the black facial cleanser from different viewpoints (front, left, back, and right), with each camera aligned with room walls and partially capturing the surroundings: If I am standing at the same spot and... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_among_group587_q1_2_2",
"original_data_source": "mindcube",
"source_subset": "mindcube/mental_simulation",
"tool_label": "mixed",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nBased on these three images (image 1, 2, and 3) showing the same scene from different viewpoints (front, left, and right): From the viewpoint presented in image 2, what is the nearest object the nearest object behind the vending machine? A. Wall B. Potted plant C. Trash ... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "mindcube_aroundnew_group065_q2_1",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a orange circle in the first image. After changing the camera or lighting, the second image shows multiple orange-circled points labeled 'A, B, C, D'. Which matches?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</thin... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_707dd634-8455-40cb-a4a0-a693311214da",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThese two images (image 1 and 2) show the same scene from two different viewpoints. Image 2 was taken after turning the camera 90 degrees to the right (clockwise) from the position of image 1. Based on these two images: If I am standing at the same spot and facing the same direct... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_rotation_group023_qsecond_3",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in cyan. The camera position or lighting is altered, and a second image is captured. Multiple cyan-circled points with 'A, B, C, D' labels appear in the second image. Identify the corresponding point.\nA point-A\nB point-B\nC point-C\nD poin... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_24c8e3bb-cb7a-43d6-bd8e-5301a74ce6ee",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a point circled in blue in the first image. After changing the camera or lighting, the second image shows several blue-circled points labeled '1, 2, 3, 4'. Which one corresponds?\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>, the... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_b74f9844-1e17-48b3-98e3-89ff26c46f8b",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn the first image, a point is circled in purple. The camera position or lighting is altered, and a second image is captured. Multiple purple-circled points with 'A, B, C, D' labels appear in the second image. Identify the corresponding point.\nA point-A\nB point-B\nC point-C\nD ... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "vst_correspondence_84139c8f-c45a-4084-b7fd-213d1b1c1cc6",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image contains a point circled in cyan. After changing the camera or lighting, the second image shows multiple cyan-circled points labeled 'A, B, C, D'. Select the matching point.\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...</think>, ... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_64c1eb54-8e71-42df-8d23-374efbd01717",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nA point is highlighted with a gold circle in image one. After adjusting the camera or lighting, image two presents multiple gold-circled points labeled 'A, B, C, D'. Which matches the original point?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>...... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_75ac4466-78a4-4caa-b8a0-a5c9a0176636",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nBased on these two views showing the same scene: in which direction did I move from the first view to the second view? A. Diagonally forward and right B. Diagonally forward and left C. Directly right D. Directly left\nAnswer with one option letter.\n\nReason step by step in <thin... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "mindcube_among_group535_q3_1_2",
"original_data_source": "mindcube",
"source_subset": "mindcube/cognitive_mapping",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nYou see a cyan circled point in image one. After changing the camera angle or lighting, image two is captured. Several points are circled in cyan and labeled with 'A, B, C, D'. Which point corresponds to the original?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by ... | spatial_reasoning | {
"ground_truth": "D",
"style": "rule"
} | {
"id": "vst_correspondence_1e12165a-5f7c-479a-8e05-9981992ea223",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nIn image one, a point is circled in cyan. The camera or lighting is changed, and image two contains multiple cyan-circled points labeled '1, 2, 3, 4'. Identify the matching point.\nA point-1\nB point-2\nC point-3\nD point-4\n\nReason step by step in <think>...</think>, then outpu... | spatial_reasoning | {
"ground_truth": "A",
"style": "rule"
} | {
"id": "vst_correspondence_d9dc1399-7458-4cf9-af09-2d1aa422c63c",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\nThe first image contains a blue-circled point. After adjusting the camera or lighting, the second image displays several blue-circled points with 'A, B, C, D' labels. Which is the corresponding point?\nA point-A\nB point-B\nC point-C\nD point-D\n\nReason step by step in <think>..... | spatial_reasoning | {
"ground_truth": "C",
"style": "rule"
} | {
"id": "vst_correspondence_d84e5797-3f3a-4e68-9e4c-595881b315a2",
"original_data_source": "vst_correspondence",
"source_subset": "vst_correspondence/mi_correspondence",
"tool_label": "likely_tool",
"train_style": "cot"
} | |
hiyouga/geometry3k | [
{
"content": "<image>\n<image>\n<image>\nThese three images (image 1, 2, and 3) show the same scene from three different viewpoints. The image 2 was taken after turning the camera 90 degrees to the right (clockwise) from the position of image 1. For image 3, the camera was turned another 90 degrees right, so it... | spatial_reasoning | {
"ground_truth": "B",
"style": "rule"
} | {
"id": "mindcube_rotation_group049_q3_6",
"original_data_source": "mindcube",
"source_subset": "mindcube/perspective_rotation_around",
"tool_label": "likely_tool",
"train_style": "cot"
} |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 362