Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910107203272704 |
|---|---|
| author | Fu, Chaoyou Yuan, Haozhi Dong, Yuhao Zhang, Yi-Fan Shen, Yunhang Hu, Xiaoxing Li, Xueying Su, Jinsen Long, Chengwu Xie, Xiaoyao Xie, Yongkang Zheng, Xiawu Yang, Xue Cao, Haoyu Wu, Yunsheng Liu, Ziwei Sun, Xing Shan, Caifeng He, Ran |
| author_facet | Fu, Chaoyou Yuan, Haozhi Dong, Yuhao Zhang, Yi-Fan Shen, Yunhang Hu, Xiaoxing Li, Xueying Su, Jinsen Long, Chengwu Xie, Xiaoyao Xie, Yongkang Zheng, Xiawu Yang, Xue Cao, Haoyu Wu, Yunsheng Liu, Ziwei Sun, Xing Shan, Caifeng He, Ran |
| contents | With the rapid advancement of video understanding, existing benchmarks are becoming increasingly saturated, exposing a critical discrepancy between inflated leaderboard scores and real-world model capabilities. To address this widening gap, we introduce Video-MME-v2, a comprehensive benchmark designed to rigorously evaluate the robustness and faithfulness of video understanding. To systematically evaluate model capabilities, we design a \textbf{progressive tri-level hierarchy} that incrementally increases the complexity of video comprehension, ranging from multi-point visual information aggregation, to temporal dynamics modeling, and ultimately to complex multimodal reasoning. Besides, in contrast to conventional per-question accuracy, we propose a \textbf{group-based non-linear evaluation} strategy that enforces both consistency across related queries and coherence in multi-step reasoning. It penalizes fragmented or guess-based correctness and assigns credit only to answers supported by valid reasoning. To guarantee data quality, Video-MME-v2 is constructed through a rigorously controlled human annotation pipeline, involving 12 annotators and 50 independent reviewers. Backed by \textbf{3,300 human-hours} and up to \textbf{5 rounds} of quality assurance, Video-MME-v2 aims to serve as one of the most authoritative video benchmarks. Extensive experiments reveal a substantial gap between current best model Gemini-3-Pro and human experts, and uncover a clear hierarchical bottleneck where errors in visual information aggregation and temporal modeling propagate to limit high-level reasoning. We further find that thinking-based reasoning is highly dependent on textual cues, improving performance with subtitles but sometimes degrading it in purely visual settings. By exposing these limitations, Video-MME-v2 establishes a demanding new testbed for the development of next-generation video MLLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_05015 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding Fu, Chaoyou Yuan, Haozhi Dong, Yuhao Zhang, Yi-Fan Shen, Yunhang Hu, Xiaoxing Li, Xueying Su, Jinsen Long, Chengwu Xie, Xiaoyao Xie, Yongkang Zheng, Xiawu Yang, Xue Cao, Haoyu Wu, Yunsheng Liu, Ziwei Sun, Xing Shan, Caifeng He, Ran Computer Vision and Pattern Recognition With the rapid advancement of video understanding, existing benchmarks are becoming increasingly saturated, exposing a critical discrepancy between inflated leaderboard scores and real-world model capabilities. To address this widening gap, we introduce Video-MME-v2, a comprehensive benchmark designed to rigorously evaluate the robustness and faithfulness of video understanding. To systematically evaluate model capabilities, we design a \textbf{progressive tri-level hierarchy} that incrementally increases the complexity of video comprehension, ranging from multi-point visual information aggregation, to temporal dynamics modeling, and ultimately to complex multimodal reasoning. Besides, in contrast to conventional per-question accuracy, we propose a \textbf{group-based non-linear evaluation} strategy that enforces both consistency across related queries and coherence in multi-step reasoning. It penalizes fragmented or guess-based correctness and assigns credit only to answers supported by valid reasoning. To guarantee data quality, Video-MME-v2 is constructed through a rigorously controlled human annotation pipeline, involving 12 annotators and 50 independent reviewers. Backed by \textbf{3,300 human-hours} and up to \textbf{5 rounds} of quality assurance, Video-MME-v2 aims to serve as one of the most authoritative video benchmarks. Extensive experiments reveal a substantial gap between current best model Gemini-3-Pro and human experts, and uncover a clear hierarchical bottleneck where errors in visual information aggregation and temporal modeling propagate to limit high-level reasoning. We further find that thinking-based reasoning is highly dependent on textual cues, improving performance with subtitles but sometimes degrading it in purely visual settings. By exposing these limitations, Video-MME-v2 establishes a demanding new testbed for the development of next-generation video MLLMs. |
| title | Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.05015 |