arXiv cs.CL· Sha Luo, Yogesh Prabhu, Timothy Ossowski, Kaiping Chen, Junjie Hu·· 13 小时前AI 评分31
RiskCueBench：评测视频语言模型从早期风险线索进行预判推理的基准
RiskCueBench: Benchmarking Anticipatory Reasoning from Early Risk Cues in Video-Language Models
AI 导读
研究团队推出视频理解基准 RiskCueBench，通过人工标注视频中最早出现安全隐患的“风险信号片段”，要求模型仅凭早期视觉线索预判后续风险事件，而非观看包含事故本身在内的完整视频。实验显示，当前系统在理解动态演变情境、从早期信号预判未来风险事件方面存在明显差距，凸显视频风险预测模型实际部署面临的挑战。
正文 · 原文
Abstract:With the rapid growth of video centered social media, the ability to anticipate risky events from visual data is a promising direction for ensuring public safety and preventing real world accidents. Prior work has extensively studied supervised video risk assessment across domains such as driving, protests, and natural disasters. However, many existing datasets provide models with access to the full video sequence, including the accident itself, which substantially reduces the difficulty of the task. To better reflect real world conditions, we introduce a new video understanding benchmark RiskCueBench in which videos are carefully annotated to identify a risk signal clip, defined as the earliest moment that indicates a potential safety concern. Experimental results reveal a significant gap in current systems ability to interpret evolving situations and anticipate future risky events from early visual signals, highlighting important challenges for deploying video risk prediction models in practice.
| Comments: | *updated author email in this version |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL) |
| Cite as: | arXiv:2601.03369 [cs.CV] |
| (or arXiv:2601.03369v3 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2601.03369 arXiv-issued DOI via DataCite |
Submission history
From: Sha Luo [view email]
[v1]
Tue, 6 Jan 2026 19:14:49 UTC (36,160 KB)
[v2]
Wed, 21 Jan 2026 06:11:42 UTC (36,160 KB)
[v3]
Wed, 21 Oct 2026 20:12:23 UTC (36,529 KB)
来源：arXiv cs.CL · arxiv.org