大模型对齐与安全

Auto-Tuning Safety Guardrails for Black-Box Large Language Models

为黑盒大语言模型自动调整安全防护栏

Perry Abdulkadir

机构 * University of St. Thomas（圣托马斯大学） 纠错

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

AI总结 本文提出通过超参数优化来自动调整黑盒大语言模型的安全防护栏，以提高其安全性和效率。

Comments 8 pages, 7 figures, 1 table. Work completed as part of the M.S. in Artificial Intelligence at the University of St. Thomas using publicly available models and datasets; all views and any errors are the author's own

Bilevel Optimization for Covert Memory Tampering in Heterogeneous Multi-Agent Architectures (XAMT)

针对异构多智能体架构中的隐蔽内存篡改的双层优化

Akhil Sharma, Shaikh Yaser Arafat, Jai Kumar Sharma, Ken Huang

专题命中 越狱攻击 ：safety(abstract)

AI总结 XAMT通过双层优化方法针对异构多智能体架构中的隐蔽内存篡改问题，提出了一种新的训练时威胁模型，以提升系统安全性。

Comments 10 pages, 5 figures, 4 tables. Conference-style paper (IEEEtran). Proposes unified bilevel optimization framework for covert memory poisoning attacks in heterogeneous multi-agent systems (MARL + RAG)