从Spark到Fire:在基于大语言模型的多智能体协作中建模和缓解错误级联
From Spark to Fire: Modeling and Mitigating Error Cascades in LLM-Based Multi-Agent Collaboration
AI总结:
本文提出了一种针对LLM-MAS的传播动力学模型,通过实验识别出三种漏洞类型,并引入基因图治理层以抑制误差级联,有效缓解了错误传播风险。
摘要:
基于大语言模型的多智能体系统(LLM-MAS)正被应用于复杂的协作场景。然而,其协作机制可能导致微小不准确逐渐固化为系统级的虚假共识。现有保护措施往往依赖单智能体验证或需要修改协作架构,可能削弱有效信息流。为此,本文提出了一种针对LLM-MAS的传播动力学模型,将协作抽象为有向依赖图,并提供早期风险标准以表征放大风险。通过六个主流框架的实验,识别出三种漏洞类别:级联放大、拓扑敏感性和共识惯性。进一步,我们实例化了一种攻击,即注入单个原子错误种子导致广泛失败。为此,引入基于基因图的治理层,作为消息层插件,抑制内生和外生误差放大,不改变协作架构。实验表明,该方法在至少89%的运行中阻止了最终感染,并显著减轻了微小误差的级联传播。
英文摘要:
Large Language Model-based Multi-Agent Systems (LLM-MAS) are increasingly applied to complex collaborative scenarios. However, their collaborative mechanisms may cause minor inaccuracies to gradually solidify into system-level false consensus through iteration. Such risks are difficult to trace since errors can propagate and amplify through message dependencies. Existing protections often rely on single-agent validation or require modifications to the collaboration architecture, which can weaken effective information flow and may not align with natural collaboration processes in real tasks. To address this, we propose a propagation dynamics model tailored for LLM-MAS that abstracts collaboration as a directed dependency graph and provides an early-stage risk criterion to characterize amplification risk. Through experiments on six mainstream frameworks, we identify three vulnerability classes: cascade amplification, topological sensitivity, and consensus inertia. We further instantiate an attack where injecting just a single atomic error seed leads to widespread failure. In response, we introduce a genealogy-graph-based governance layer, implemented as a message-layer plugin, that suppresses both endogenous and exogenous error amplification without altering the collaboration architecture. Experiments show that this approach prevents final infection in at least 89% of runs across operating modes and significantly mitigates the cascading spread of minor errors.
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。