HazardArena: Evaluating Semantic Safety in Vision-Language-Action Models

作者：Vision–Language–Action Models Zixing Chen, Yifeng Gao, Li Wang, Yunhan Zhao, Yi Liu, Jiayu Li, Xiang Zheng, Zuxuan Wu, Cong Wang, Xingjun Ma, Yu-Gang Jiang · 单位：Fudan University, Shanghai Innovation Institute, City University of Hong Kong · 会议/期刊：arXiv · 日期：2026-04-14 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Robustness / Attack