Datasets:
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
schema_version: string
id: string
dataset_type: string
problem_type: string
data_type: string
image_path: string
audio_path: string
correct_answer: string
messages: list<item: struct<role: string, content: list<item: string>>>
child 0, item: struct<role: string, content: list<item: string>>
child 0, role: string
child 1, content: list<item: string>
child 0, item: string
hindsight_trace: string
video_path: string
video_uid: string
duration: double
qa_id: string
stage1_mode: string
zoomin_interval: list<item: double>
child 0, item: double
gt_intervals: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
stage1_interval: list<item: double>
child 0, item: double
anomaly_type: string
to
{'schema_version': Value('string'), 'id': Value('string'), 'dataset_type': Value('string'), 'problem_type': Value('string'), 'video_path': Value('string'), 'video_uid': Value('string'), 'duration': Value('float64'), 'gt_intervals': List(List(Value('float64'))), 'stage1_mode': Value('string'), 'stage1_interval': List(Value('float64')), 'zoomin_interval': List(Value('float64')), 'correct_answer': Value('string'), 'messages': List({'role': Value('string'), 'content': List(Json(decode=True))}), 'hindsight_trace': Json(decode=True), 'data_type': Value('string'), 'image_path': Value('string'), 'audio_path': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1779, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 609, in wrapped
for item in generator(*args, **kwargs):
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 299, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
schema_version: string
id: string
dataset_type: string
problem_type: string
data_type: string
image_path: string
audio_path: string
correct_answer: string
messages: list<item: struct<role: string, content: list<item: string>>>
child 0, item: struct<role: string, content: list<item: string>>
child 0, role: string
child 1, content: list<item: string>
child 0, item: string
hindsight_trace: string
video_path: string
video_uid: string
duration: double
qa_id: string
stage1_mode: string
zoomin_interval: list<item: double>
child 0, item: double
gt_intervals: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
stage1_interval: list<item: double>
child 0, item: double
anomaly_type: string
to
{'schema_version': Value('string'), 'id': Value('string'), 'dataset_type': Value('string'), 'problem_type': Value('string'), 'video_path': Value('string'), 'video_uid': Value('string'), 'duration': Value('float64'), 'gt_intervals': List(List(Value('float64'))), 'stage1_mode': Value('string'), 'stage1_interval': List(Value('float64')), 'zoomin_interval': List(Value('float64')), 'correct_answer': Value('string'), 'messages': List({'role': Value('string'), 'content': List(Json(decode=True))}), 'hindsight_trace': Json(decode=True), 'data_type': Value('string'), 'image_path': Value('string'), 'audio_path': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1342, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 907, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1832, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
schema_version string | id string | dataset_type string | problem_type string | video_path string | video_uid string | duration float64 | gt_intervals list | stage1_mode string | stage1_interval list | zoomin_interval list | correct_answer string | messages list | hindsight_trace unknown | data_type string | image_path string | audio_path string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
open_r1.coldstart_sft.v2 | cgbench:odV0hLIP5gs::q1174 | cgbench | multiple_choice | media/videos/cgbench/odV0hLIP5gs.mp4 | odV0hLIP5gs | 1,505 | [
[
0,
16.4
]
] | interval | [
0,
16
] | [
0,
16
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1174,
"domain": "Art & Culture",
"sub_category": "2D Spatial Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/odV0hLIP5gs.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sand... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Qe411m7hF::q1201 | cgbench | multiple_choice | media/videos/cgbench/BV1Qe411m7hF.mp4 | BV1Qe411m7hF | 1,911 | [
[
1866,
1885
]
] | interval | [
1866,
1885
] | [
1866,
1885
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1201,
"domain": "Music & TV show",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Qe411m7hF.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1vg4y157v5::q1162 | cgbench | multiple_choice | media/videos/cgbench/BV1vg4y157v5.mp4 | BV1vg4y157v5 | 1,616 | [
[
203,
213
]
] | interval | [
203,
213
] | [
203,
213
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1162,
"domain": "Special Scenes",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1vg4y157v5.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:k5uns2AY-Qs::q1151 | cgbench | multiple_choice | media/videos/cgbench/k5uns2AY-Qs.mp4 | k5uns2AY-Qs | 1,738 | [
[
1128,
1136
]
] | interval | [
1120,
1150
] | [
1120,
1150
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1151,
"domain": "Special Scenes",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/k5uns2AY-Qs.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:G3GMeH5S6hA::q1166 | cgbench | multiple_choice | media/videos/cgbench/G3GMeH5S6hA.mp4 | G3GMeH5S6hA | 1,283 | [
[
400,
430
]
] | interval | [
400,
430
] | [
400,
430
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1166,
"domain": "Life Record",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/G3GMeH5S6hA.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_ca... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1eU421d71N::q1116 | cgbench | multiple_choice | media/videos/cgbench/BV1eU421d71N.mp4 | BV1eU421d71N | 2,286 | [
[
2115,
2127
]
] | interval | [
2115,
2127
] | [
2115,
2127
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1116,
"domain": "Music & TV show",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1eU421d71N.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1by421b7Bm::q1141 | cgbench | multiple_choice | media/videos/cgbench/BV1by421b7Bm.mp4 | BV1by421b7Bm | 1,115 | [
[
50,
93.4
]
] | interval | [
50,
93
] | [
50,
93
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1141,
"domain": "Music & TV show",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1by421b7Bm.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1cX4y1m7h2::q1218 | cgbench | multiple_choice | media/videos/cgbench/BV1cX4y1m7h2.mp4 | BV1cX4y1m7h2 | 1,652 | [
[
162.7,
441.4
]
] | interval | [
162.7,
441.4
] | [
162.7,
441.4
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1218,
"domain": "Music & TV show",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1cX4y1m7h2.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Z1421C77m::q1241 | cgbench | multiple_choice | media/videos/cgbench/BV1Z1421C77m.mp4 | BV1Z1421C77m | 1,053 | [
[
15,
20
]
] | interval | [
15,
25
] | [
15,
25
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1241,
"domain": "Electonic/Social Gaming",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Z1421C77m.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV15w411s7kE::q1246 | cgbench | multiple_choice | media/videos/cgbench/BV15w411s7kE.mp4 | BV15w411s7kE | 1,264 | [
[
618.8,
636.9
]
] | interval | [
618.8,
636.9
] | [
618.8,
636.9
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1246,
"domain": "Art & Culture",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV15w411s7kE.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:4d1X2AxnZ4o::q1291 | cgbench | multiple_choice | media/videos/cgbench/4d1X2AxnZ4o.mp4 | 4d1X2AxnZ4o | 1,221 | [
[
212,
218
]
] | interval | [
210,
230
] | [
210,
230
] | A | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1291,
"domain": "Art & Culture",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/4d1X2AxnZ4o.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:fkKrGUwqZ40::q1236 | cgbench | multiple_choice | media/videos/cgbench/fkKrGUwqZ40.mp4 | fkKrGUwqZ40 | 6,325 | [
[
15,
23
]
] | interval | [
15,
25
] | [
15,
25
] | C | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1236,
"domain": "Music & TV show",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/fkKrGUwqZ40.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:CDYJME8fgkI::q1220 | cgbench | multiple_choice | media/videos/cgbench/CDYJME8fgkI.mp4 | CDYJME8fgkI | 3,236 | [
[
325.2,
328.7
]
] | interval | [
320,
340
] | [
320,
340
] | C | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1220,
"domain": "Music & TV show",
"sub_category": "Text Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/CDYJME8fgkI.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1UV411W7He::q1260 | cgbench | multiple_choice | media/videos/cgbench/BV1UV411W7He.mp4 | BV1UV411W7He | 1,851 | [
[
60,
70
]
] | interval | [
60,
70
] | [
60,
70
] | H | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1260,
"domain": "Life Record",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1UV411W7He.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cac... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1k84y1f7KG::q1337 | cgbench | multiple_choice | media/videos/cgbench/BV1k84y1f7KG.mp4 | BV1k84y1f7KG | 1,653 | [
[
1439.3,
1474.6
]
] | interval | [
1439,
1475
] | [
1439,
1475
] | H | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1337,
"domain": "Life Record",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1k84y1f7KG.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cac... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Mx4y1S7Tf::q1523 | cgbench | multiple_choice | media/videos/cgbench/BV1Mx4y1S7Tf.mp4 | BV1Mx4y1S7Tf | 1,226 | [
[
250,
280
]
] | interval | [
250,
280
] | [
250,
280
] | B | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1523,
"domain": "Electonic/Social Gaming",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Mx4y1S7Tf.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:x8LBUDE8c4A::q1409 | cgbench | multiple_choice | media/videos/cgbench/x8LBUDE8c4A.mp4 | x8LBUDE8c4A | 2,883 | [
[
60,
70
]
] | interval | [
60,
70
] | [
60,
70
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1409,
"domain": "Art & Culture",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/x8LBUDE8c4A.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:tBWMn4y1Yfo::q1542 | cgbench | multiple_choice | media/videos/cgbench/tBWMn4y1Yfo.mp4 | tBWMn4y1Yfo | 2,622 | [
[
597,
600
]
] | interval | [
590,
610
] | [
590,
610
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1542,
"domain": "GUI",
"sub_category": "Text Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/tBWMn4y1Yfo.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cache/stage1... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:PZ0NQoAcqfc::q1564 | cgbench | multiple_choice | media/videos/cgbench/PZ0NQoAcqfc.mp4 | PZ0NQoAcqfc | 3,036 | [
[
540,
570
]
] | interval | [
540,
570
] | [
540,
570
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1564,
"domain": "Music & TV show",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/PZ0NQoAcqfc.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:MAuVJ880jv0::q1604 | cgbench | multiple_choice | media/videos/cgbench/MAuVJ880jv0.mp4 | MAuVJ880jv0 | 1,564 | [
[
60,
72
]
] | interval | [
60,
72
] | [
60,
72
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1604,
"domain": "Art & Culture",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/MAuVJ880jv0.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Ru4y1J7Ee::q1408 | cgbench | multiple_choice | media/videos/cgbench/BV1Ru4y1J7Ee.mp4 | BV1Ru4y1J7Ee | 1,314 | [
[
27,
57
]
] | interval | [
27,
57
] | [
27,
57
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1408,
"domain": "Music & TV show",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Ru4y1J7Ee.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandb... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1LJn9epE7D::q1578 | cgbench | multiple_choice | media/videos/cgbench/BV1LJn9epE7D.mp4 | BV1LJn9epE7D | 1,837 | [
[
607,
614
]
] | interval | [
600,
620
] | [
600,
620
] | A | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1578,
"domain": "Electonic/Social Gaming",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1LJn9epE7D.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1K94y1X7Ca::q1770 | cgbench | multiple_choice | media/videos/cgbench/BV1K94y1X7Ca.mp4 | BV1K94y1X7Ca | 1,166 | [
[
115,
120
]
] | interval | [
110,
140
] | [
110,
140
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1770,
"domain": "Humor/funny",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1K94y1X7Ca.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cac... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:xJ5Ia7hbY9o::q1744 | cgbench | multiple_choice | media/videos/cgbench/xJ5Ia7hbY9o.mp4 | xJ5Ia7hbY9o | 1,668 | [
[
1002,
1004
]
] | interval | [
1000,
1010
] | [
1000,
1010
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1744,
"domain": "Life Record",
"sub_category": "Text Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/xJ5Ia7hbY9o.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cach... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:7ip-qsC0Ybc::q1798 | cgbench | multiple_choice | media/videos/cgbench/7ip-qsC0Ybc.mp4 | 7ip-qsC0Ybc | 1,415 | [
[
169.1,
188.5
]
] | interval | [
169.1,
188.5
] | [
169.1,
188.5
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1798,
"domain": "Art & Culture",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/7ip-qsC0Ybc.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Jm421K7hc::q1841 | cgbench | multiple_choice | media/videos/cgbench/BV1Jm421K7hc.mp4 | BV1Jm421K7hc | 1,378 | [
[
32.3,
44.4
]
] | interval | [
32.3,
44.4
] | [
32.3,
44.4
] | B | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1841,
"domain": "Special Scenes",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Jm421K7hc.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1tu4y187gY::q1959 | cgbench | multiple_choice | media/videos/cgbench/BV1tu4y187gY.mp4 | BV1tu4y187gY | 1,531 | [
[
530,
540
]
] | interval | [
530,
540
] | [
530,
540
] | B | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1959,
"domain": "Electonic/Social Gaming",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1tu4y187gY.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1UY4y1Q7WP::q1915 | cgbench | multiple_choice | media/videos/cgbench/BV1UY4y1Q7WP.mp4 | BV1UY4y1Q7WP | 3,197 | [
[
135,
360
]
] | interval | [
135,
360
] | [
135,
360
] | A | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1915,
"domain": "Instruction & Knowledge",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1UY4y1Q7WP.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1tz421Y7Vq::q1964 | cgbench | multiple_choice | media/videos/cgbench/BV1tz421Y7Vq.mp4 | BV1tz421Y7Vq | 2,584 | [
[
1235,
1320
]
] | interval | [
1235,
1320
] | [
1235,
1320
] | B | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1964,
"domain": "Art & Culture",
"sub_category": "Scene Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1tz421Y7Vq.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Ma4y1u7cr::q1593 | cgbench | multiple_choice | media/videos/cgbench/BV1Ma4y1u7cr.mp4 | BV1Ma4y1u7cr | 2,160 | [
[
295,
303
]
] | interval | [
290,
310
] | [
290,
310
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1593,
"domain": "Special Scenes",
"sub_category": "Text Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Ma4y1u7cr.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:d_vFKszLtik::q2008 | cgbench | multiple_choice | media/videos/cgbench/d_vFKszLtik.mp4 | d_vFKszLtik | 1,624 | [
[
985,
992
]
] | interval | [
980,
1000
] | [
980,
1000
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2008,
"domain": "Art & Culture",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/d_vFKszLtik.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:d_vFKszLtik::q1980 | cgbench | multiple_choice | media/videos/cgbench/d_vFKszLtik.mp4 | d_vFKszLtik | 1,624 | [
[
1227,
1232
]
] | interval | [
1220,
1240
] | [
1220,
1240
] | A | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1980,
"domain": "Art & Culture",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/d_vFKszLtik.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1MP411D7H8::q2070 | cgbench | multiple_choice | media/videos/cgbench/BV1MP411D7H8.mp4 | BV1MP411D7H8 | 1,293 | [
[
95.9,
143.2
]
] | interval | [
95,
145
] | [
95,
145
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2070,
"domain": "Art & Culture",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1MP411D7H8.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:_Xnkq0ZjBXc::q2045 | cgbench | multiple_choice | media/videos/cgbench/_Xnkq0ZjBXc.mp4 | _Xnkq0ZjBXc | 1,759 | [
[
240,
252
]
] | interval | [
240,
252
] | [
240,
252
] | C | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2045,
"domain": "Art & Culture",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/_Xnkq0ZjBXc.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_c... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV13i421Q7Do::q2112 | cgbench | multiple_choice | media/videos/cgbench/BV13i421Q7Do.mp4 | BV13i421Q7Do | 3,178 | [
[
2524,
2528
]
] | interval | [
2510,
2540
] | [
2510,
2540
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2112,
"domain": "Life Record",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV13i421Q7Do.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_ca... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1ae4y1d78n::q1620 | cgbench | multiple_choice | media/videos/cgbench/BV1ae4y1d78n.mp4 | BV1ae4y1d78n | 3,526 | [
[
217,
221
]
] | interval | [
210,
230
] | [
210,
230
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 1620,
"domain": "Special Scenes",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1ae4y1d78n.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbo... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:PAA_lCutsfE::q2086 | cgbench | multiple_choice | media/videos/cgbench/PAA_lCutsfE.mp4 | PAA_lCutsfE | 3,103 | [
[
290,
295
]
] | interval | [
280,
310
] | [
280,
310
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2086,
"domain": "GUI",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/PAA_lCutsfE.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cache/stage1... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:eQNSD8GAOtE::q2016 | cgbench | multiple_choice | media/videos/cgbench/eQNSD8GAOtE.mp4 | eQNSD8GAOtE | 1,361 | [
[
499,
503
]
] | interval | [
490,
510
] | [
490,
510
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2016,
"domain": "Art & Culture",
"sub_category": "Entity Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/eQNSD8GAOtE.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:AneAgyUxBsg::q2138 | cgbench | multiple_choice | media/videos/cgbench/AneAgyUxBsg.mp4 | AneAgyUxBsg | 3,409 | [
[
646,
651
]
] | interval | [
630,
660
] | [
630,
660
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2138,
"domain": "Life Record",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/AneAgyUxBsg.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cach... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1Nc41157Gx::q2211 | cgbench | multiple_choice | media/videos/cgbench/BV1Nc41157Gx.mp4 | BV1Nc41157Gx | 771 | [
[
11.9,
28.8
]
] | interval | [
11.9,
28.8
] | [
11.9,
28.8
] | F | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2211,
"domain": "Music & TV show",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1Nc41157Gx.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:EsfjaoGRXtU::q2212 | cgbench | multiple_choice | media/videos/cgbench/EsfjaoGRXtU.mp4 | EsfjaoGRXtU | 1,737 | [
[
1335,
1343
]
] | interval | [
1330,
1360
] | [
1330,
1360
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2212,
"domain": "Music & TV show",
"sub_category": "Text Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/EsfjaoGRXtU.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:8SncSXQM_x4::q2238 | cgbench | multiple_choice | media/videos/cgbench/8SncSXQM_x4.mp4 | 8SncSXQM_x4 | 1,617 | [
[
954,
960
]
] | interval | [
940,
970
] | [
940,
970
] | D | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2238,
"domain": "Life Record",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/8SncSXQM_x4.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cac... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV1nJ4m1V7NR::q2233 | cgbench | multiple_choice | media/videos/cgbench/BV1nJ4m1V7NR.mp4 | BV1nJ4m1V7NR | 1,182 | [
[
151,
156
]
] | interval | [
145,
165
] | [
145,
165
] | E | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2233,
"domain": "Electonic/Social Gaming",
"sub_category": "Entity Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV1nJ4m1V7NR.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/ful... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:BV16q4y1z7ya::q2131 | cgbench | multiple_choice | media/videos/cgbench/BV16q4y1z7ya.mp4 | BV16q4y1z7ya | 3,502 | [
[
540,
600
]
] | interval | [
540,
600
] | [
540,
600
] | A | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2131,
"domain": "Music & TV show",
"sub_category": "Event Cognition",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/BV16q4y1z7ya.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox... | null | null | null |
open_r1.coldstart_sft.v2 | cgbench:XtQMytORBmM::q2288 | cgbench | multiple_choice | media/videos/cgbench/XtQMytORBmM.mp4 | XtQMytORBmM | 2,799 | [
[
4,
10
]
] | interval | [
0,
30
] | [
0,
30
] | G | [
{
"role": "system",
"content": [
{
"type": "text",
"text": "You are an audio-visual assistant.\n\nOutput tags only, with no text outside tags.\nUse exactly one:\n<think>...</think><interval>[start, end]</interval>\nor\n<think>...</think><answer>...</answer>\nNever output both <interval... | {
"source": {
"qid": 2288,
"domain": "GUI",
"sub_category": "Event Perception",
"required_modality": "audio_visual"
},
"teacher_media": {
"muted_full_video": "media/videos/cgbench/XtQMytORBmM.mp4",
"stage1_video": "data_factory/sft-cot/output/cgbench_hindsight_full/full.sandbox_cache/stage... | null | null | null |
End of preview.
OmniReasoner-SFT
OmniReasoner-SFT is a mixed-source, research-only supervised fine-tuning dataset for audio-visual and long-video reasoning. It contains two-stage cold-start SFT trajectories with interval selection, zoom-in evidence, and final answers.
Contents
data/train.jsonl: HF-ready training JSONL with repo-relative media paths.media/: raw and derived media referenced bytrain.jsonl.manifests/media_manifest.jsonl: media inventory with repo paths, source family, media type, file size, and reference counts.manifests/dataset_stats.json: dataset and media statistics.configs/: lmms-engine SFT configs and launch scripts used for training.scripts/materialize_lmms_jsonl.py: rewrites repo-relative paths into local absolutefile://paths after download.
Source Families
This dataset includes both original-source and derived media:
- CG-Bench long videos.
- FineVideo videos.
- AVQA-R1 image/audio samples derived from OmniInstruct/AVQA/VGGSound sources.
- Composition videos generated by concatenating multiple FineVideo videos.
- Anomaly videos generated by inserting LLaVA-Video-178K / scaled-source anomaly clips into CG-Bench long videos, with anomaly type and ground-truth intervals.
The final SFT trajectories were produced with Gemini hindsight generation and then filtered with leakage checks, hard rules, judge scoring, and fusion.
License And Usage
The annotations, reasoning traces, metadata, and construction recipes are released by the dataset authors for non-commercial academic research only.
Raw and derived media retain the copyright and usage restrictions of their source datasets, original creators, and original platforms. Access to this repository does not grant additional rights beyond those source licenses and terms. Users must not redistribute the raw or derived media.
If you are a rights holder and want a file removed, please contact the dataset authors through the Hugging Face repository discussion page.
Citations And Source Notices
Users should cite this dataset and the applicable upstream datasets, including:
- CG-Bench: Clue-grounded Question Answering Benchmark for Long Video Understanding.
- FineVideo and YouTube-Commons where applicable.
- LLaVA-Video-178K for anomaly inserted-source clips where applicable.
- AVQA-R1 / EchoInk-R1, OmniInstruct/OmniBench, AVQA, and VGGSound where applicable.
Local Training
data/train.jsonl stores relative paths such as media/videos/.... To recreate
the absolute file:// format expected by the original lmms-engine training
environment after downloading this repository, run:
python scripts/materialize_lmms_jsonl.py \
--input data/train.jsonl \
--output data/train.local.jsonl \
--root "$(pwd)"
Then point the lmms-engine config to data/train.local.jsonl.
- Downloads last month
- 12,059