arXiv cs.AI· Yuqi Jia, Hangsheng Zhang, Chenxi Liu, Junjie Liang, Xun Chen·· 13 小时前AI 评分39
ReMAS：用强化学习对多智能体系统做黑盒红队测试
Black-Box Red-Teaming of Multi-Agent Systems via Reinforcement Learning
AI 导读
研究者提出 ReMAS，首个面向多智能体系统（MAS）的强化学习红队框架，通过微调攻击者 LLM 生成对抗提示词以提取系统提示词。该框架分两步：改写阶段优化基础攻击提示词，模板生成阶段构造更易调用特定子智能体的攻击模板。实验显示其攻击成功率显著高于现有黑盒攻击方法，并可跨不同后端 LLM 与 MAS 结构迁移。
正文 · 原文
Abstract:Large language model (LLM) agents are increasingly deployed in multi-agent systems (MAS) to accomplish complex tasks. Prior black-box red-teaming attacks mainly focus on single agents, but we find that these methods are far less effective in MAS, where multiple sub-agents may not directly interact with the user. Therefore, we introduce ReMAS, the first reinforcement learning-based red-teaming framework tailored to MASs, which fine-tunes attacker LLMs to generate effective adversarial prompts for system prompt extraction. The framework follows a two-step process: first, a rewrite stage refines base attack prompts to increase extraction success, and second, a template generation stage constructs attack templates that improve the likelihood of invoking specific sub-agents and thus revealing their system prompts. Extensive experiments show that our method substantially improves attack success rates compared to existing black-box attacks, with transferability across different backend LLMs and MAS structures. These results underscore the vulnerabilities of multi-agent systems and the importance of developing stronger defenses against adaptive black-box adversaries.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.28966 [cs.CR] |
| (or arXiv:2610.28966v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28966 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuqi Jia [view email]
[v1]
Tue, 20 Oct 2026 22:32:22 UTC (212 KB)
来源：arXiv cs.AI · arxiv.org