超大规模视频推理套件

摘要

视频模型的快速发展主要聚焦于视觉质量，其推理能力尚未得到充分探索。视频推理将智能根植于时空一致的视觉环境中，这种环境超越了文本自然捕捉的范畴，能够实现对连续性、交互性和因果性等时空结构的直观推理。然而，由于缺乏大规模训练数据，系统研究视频推理及其扩展规律面临挑战。为填补这一空白，我们推出了超大规模视频推理数据集（VBVR），该资源涵盖200个遵循原理化分类法的精选推理任务，包含超过100万个视频片段，规模较现有数据集提升约三个数量级。我们进一步推出VBVR-Bench可验证评估框架，通过引入基于规则且与人类判断对齐的评分机制，超越基于模型的评判方式，实现对视频推理能力的可复现、可解释诊断。借助VBVR套件，我们开展了首批大规模视频推理扩展研究，观察到模型对未见推理任务出现早期涌现泛化迹象。VBVR为可泛化视频推理的下一阶段研究奠定了基础。数据、基准工具包及模型已公开于https://video-reason.com/。

English

Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally consistent visual environments that go beyond what text can naturally capture, enabling intuitive reasoning over spatiotemporal structure such as continuity, interaction, and causality. However, systematically studying video reasoning and its scaling behavior is hindered by the lack of large-scale training data. To address this gap, we introduce the Very Big Video Reasoning (VBVR) Dataset, an unprecedentedly large-scale resource spanning 200 curated reasoning tasks following a principled taxonomy and over one million video clips, approximately three orders of magnitude larger than existing datasets. We further present VBVR-Bench, a verifiable evaluation framework that moves beyond model-based judging by incorporating rule-based, human-aligned scorers, enabling reproducible and interpretable diagnosis of video reasoning capabilities. Leveraging the VBVR suite, we conduct one of the first large-scale scaling studies of video reasoning and observe early signs of emergent generalization to unseen reasoning tasks. Together, VBVR lays a foundation for the next stage of research in generalizable video reasoning. The data, benchmark toolkit, and models are publicly available at https://video-reason.com/ .

超大规模视频推理套件

A Very Big Video Reasoning Suite

摘要

Support