zx2556/Video-MME-sampled-short

---
license: cc-by-nc-sa-4.0
task_categories:
- video-text-to-text
- visual-question-answering
language:
- en
tags:
- video
- benchmark
- video-mme
size_categories:
- n<1K
---

# Video-MME — Long videos, first 6 sub-categories

A filtered subset of [lmms-lab/Video-MME](https://huggingface.co/datasets/lmms-lab/Video-MME)
containing only:

- `duration == "long"`
- `sub_category` in:
- Humanity & History
- Literature & Art
- Biology & Medicine
- Finance & Commerce
- Astronomy
- Geography

| Field | Type |
|----------------|-------------|
| video_id | string |
| duration | string |
| domain | string |
| sub_category | string |
| url | string (YouTube link) |
| videoID | string |
| question_id | string |
| task_type | string |
| question | string |
| options | list[string] (4 choices) |
| answer | string (A/B/C/D) |

The video files themselves are **not** redistributed here — only the
question/answer metadata, mirroring the original dataset. Use the `url` /
`videoID` fields to fetch the source videos.

## Citation

```bibtex
@article{fu2024video,
title={Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis},
author={Fu, Chaoyou and Dai, Yuhan and Luo, Yondong and Li, Lei and Ren, Shuhuai and Zhang, Renrui and Wang, Zihan and Zhou, Chenyu and Shen, Yunhang and Zhang, Mengdan and others},
journal={arXiv preprint arXiv:2405.21075},
year={2024}
}
```