CORBA:针对基于大语言模型的多智能体系统的传染性递归阻断攻击
CORBA: Contagious Recursive Blocking Attacks on Multi-Agent Systems Based on Large Language Models
AI总结:
提出 Corba 攻击,利用传染性与递归性跨拓扑传播并持续消耗资源,揭示 AutoGen、Camel 等 LLM 多智能体系统在商业与开源模型下的安全脆弱性。
摘要:
基于大语言模型的多智能体系统(LLM-MAS)已展现出卓越的现实能力,能够有效协作完成复杂任务。尽管这些系统设计有安全机制,例如通过对齐拒绝有害指令,但其安全性在很大程度上仍未得到探索。这一空白使 LLM-MAS 容易受到有针对性的破坏。本文提出传染性递归阻断攻击(Contagious Recursive Blocking Attacks,Corba),这是一种新颖、简单但高效的攻击,可破坏 LLM-MAS 内智能体之间的交互。Corba 利用两个关键特性:其传染性使其能够跨越任意网络拓扑传播,而递归特性使其能够持续耗尽计算资源。值得注意的是,这些阻断攻击通常涉及看似良性的指令,因此使用传统对齐方法尤其难以缓解。我们在 AutoGen 和 Camel 两个广泛使用的 LLM-MAS 上,针对多种拓扑和商业模型评估了 Corba。此外,我们在开放式交互式 LLM-MAS 中进行了更广泛的实验,证明了 Corba 在复杂拓扑结构和开源模型中的有效性。代码可在 https://github.com/zhrli324/Corba 获取。
英文摘要:
Large Language Model-based Multi-Agent Systems (LLM-MASs) have demonstrated remarkable real-world capabilities, effectively collaborating to complete complex tasks. While these systems are designed with safety mechanisms, such as rejecting harmful instructions through alignment, their security remains largely unexplored. This gap leaves LLM-MASs vulnerable to targeted disruptions. In this paper, we introduce Contagious Recursive Blocking Attacks (Corba), a novel and simple yet highly effective attack that disrupts interactions between agents within an LLM-MAS. Corba leverages two key properties: its contagious nature allows it to propagate across arbitrary network topologies, while its recursive property enables sustained depletion of computational resources. Notably, these blocking attacks often involve seemingly benign instructions, making them particularly challenging to mitigate using conventional alignment methods. We evaluate Corba on two widely-used LLM-MASs, namely, AutoGen and Camel across various topologies and commercial models. Additionally, we conduct more extensive experiments in open-ended interactive LLM-MASs, demonstrating the effectiveness of Corba in complex topology structures and open-source models. Our code is available at: https://github.com/zhrli324/Corba.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。