论文

arXiv 论文提出 ReMA 攻击与 SecMAS 防御,加固多智能体系统安全

Securing Multi-Agent Systems by Design

精选理由

搞多智能体系统的开发者该看看这篇:恶意 agent 能靠迭代优化的响应操纵最终答案,论文给的 SecMAS 聚合规则靠语义相似度筛掉异常响应,理论上有安全证明,实验里 utility 也几乎不掉。

一篇 arXiv 论文系统研究了多智能体系统的安全漏洞。作者提出响应操纵攻击(ReMA),将无目标与有目标攻击形式化为优化问题,通过代理多智能体系统的反馈迭代优化恶意响应,逐步放大攻击效果。

正文

防御侧提出的 SecMAS 是一种安全聚合规则,利用各智能体响应之间的语义相似度,选出与足够大的响应子集最一致的那条响应作为最终输出。论文从理论上证明了 SecMAS 对无目标与有目标两类攻击的安全保证。

实验部分显示,现有聚合规则在 ReMA 下普遍失效,而 SecMAS 能挡住包括自适应变体在内的全部受测攻击,同时效用与现有聚合规则相当。对部署多智能体系统的团队来说,这给出了一个可替换的聚合层方案。

原文 · arXiv cs.AI

Securing Multi-Agent Systems by Design

Multi-agent systems solve complex tasks by decomposing them into subtasks, delegating the subtasks to agents, and aggregating the agents' responses using an aggregation rule. Malicious agents can manipulate the final solution by providing malicious responses to their assigned subtasks. In this paper, we first propose response manipulation attacks (ReMA), which iteratively optimize malicious responses to maximize attack effectiveness. Specifically, we formulate untargeted and targeted attacks as optimization problems and develop an approach to approximately solve these problems by iteratively refining malicious responses based on feedback from a surrogate multi-agent system. To defend against such attacks, we further propose SecMAS, a secure aggregation rule that leverages the semantic similarities among agent responses to select the response that is most consistent with a sufficiently large subset of the other responses. We theoretically establish security guarantees for SecMAS against both untargeted and targeted attacks. Extensive evaluations demonstrate that existing aggregation rules are vulnerable to ReMA, whereas SecMAS effectively defends against all evaluated attacks, including adaptive variants of ReMA, while maintaining utility comparable to that of existing aggregation rules.

产业坐标与信源核验

本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:多智能体、ReMA、SecMAS、AI安全,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Xilong Wang, Xiaoxing Chen, Neil Gong),经 AITOP 多维评分管线结构化提炼。