HuggingFace Daily Papers（社区热门论文）· HuggingFace Daily Papers（社区热门论文）·· 2026-05-11精选AI 评分70
WorldReasonBench：面向未来世界状态预测的视频生成器人类对齐压力测试
AI 导读
研究团队发布WorldReasonBench基准，旨在直接评估视频生成模型作为“世界模拟器”的推理能力。该基准包含436个测试案例，涵盖物理、社会、逻辑和信息四大维度及22个子类，要求模型根据初始状态与动作生成状态演化一致的未来视频。评估采用人类对齐的双部分方法：过程感知推理验证通过结构化问答检测时序与因果错误；多维质量评估则对推理质量、时序一致性和视觉美学进行评分。测试发现，当前先进模型在视觉合理性与世界推理能力间存在显著差距，生成的视频可能看似逼真却违反动态、因果或信息守恒规律。相关资源已开源。
推荐理由
视频生成越来越像真的，但逻辑和因果一塌糊涂，这个基准把问题量化了，想做世界模拟器的团队可以拿来测测自己的模型到底懂不懂世界。
正文
Authors:Keming Wu, Yijing Cui, Wenhan Xue, Qijie Wang, Xuan Luo, Zhiyuan Feng, Zuhao Yang, Sudong Wang, Sicong Jiang, Haowei Zhu, Zihan Wang, Ping Nie, Wenhu Chen, Bin Wang
Abstract:Commercial video generation systems such as Seedance2.0 and Veo3.1 have rapidly improved, strengthening the view that video generators may be evolving into "world simulators." Yet the community still lacks a benchmark that directly tests whether a model can reason about how an observed world should evolve over time. We introduce WorldReasonBench, which reframes video generation evaluation as world-state prediction: given an initial state and an action, can a model generate a future video whose state evolution remains physically, socially, logically, and informationally consistent? WorldReasonBench contains 436 curated test cases with structured ground-truth QA annotations spanning four reasoning dimensions and 22 subcategories. We evaluate generated videos with a human-aligned two-part methodology: Process-aware Reasoning Verification uses structured QA and reasoning-phase diagnostics to detect temporal and causal failures, while Multi-dimensional Quality Assessment scores reasoning quality, temporal consistency, and visual aesthetics for ranking and reward modeling. We further introduce WorldRewardBench, a preference benchmark with approximately 6K expert-annotated pairs over 1.4K videos, supporting pair-wise and point-wise reward-model evaluation. Across modern video generators, our results expose a persistent gap between visual plausibility and world reasoning: videos can look convincing while failing dynamics, causality, or information preservation. We will release our benchmarks and evaluation toolkit to support community research on genuinely world-aware video generation at this https URL.
| Comments: | Project Page: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2605.10434 [cs.CV] |
| (or arXiv:2605.10434v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.10434 arXiv-issued DOI via DataCite |
Submission history
From: Keming Wu [view email]
[v1]
Mon, 11 May 2026 12:06:57 UTC (37,590 KB)
来源：HuggingFace Daily Papers（社区热门论文） · arxiv.org