大模型对齐与安全

Noise Injection Systemically Degrades Large Language Model Safety Guardrails

Prithviraj Singh Shahani, Kaveh Eskandari Miandoab, Matthias Scheutz

机构 * Tufts University（塔夫茨大学） 纠错

专题命中 越狱攻击 ：safety(title,abstract)；alignment(abstract)；AI safety(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 9 pages,3 figures

VisualDAN: Exposing Vulnerabilities in VLMs with Visual-Driven DAN Commands

Aofan Liu, Lulu Tang

机构 * Beijing Academy of Artificial Intelligence（北京人工智能研究院） 纠错

专题命中 越狱攻击 ：alignment(abstract)；jailbreak(abstract)；分类 cs.AI

DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent

Pengyu Zhu, Zhenhong Zhou, Yuanhe Zhang, Shilinlu Yan, Kun Wang, Sen Su

机构 * Beijing University of Posts and Telecommunications（北京邮电大学） ； Nanyang Technological University（南洋理工大学） 纠错

专题命中 越狱攻击 ：safety(abstract)；分类 cs.AI

Collaborative Shadows: Distributed Backdoor Attacks in LLM-Based Multi-Agent Systems

Pengyu Zhu, Lijun Li, Yaxing Lyu, Li Sun, Sen Su, Jing Shao

专题命中 越狱攻击 ：safety(abstract)