将大模型多智能体系统的日志转化为因果图，精准定位故障根源。
From Flat Logs to Causal Graphs: Hierarchical Failure Attribution for LLM-based Multi-Agent Systems
- 把线性日志转为分层因果图，理清智能体间的依赖关系。
- 在Who&When基准上，任务级和步骤级准确率均超越8个顶尖基线。
- 适合研究多智能体系统可解释性与故障诊断的开发者使用。
基于大语言模型的多智能体系统在复杂任务中表现出色，但存在固有的脆弱性和故障机制不透明的问题。现有故障归因方法依赖直接提示、昂贵重放或监督微调，通常将执行日志视为线性序列，无法拆解多智能体系统中复杂的因果关系，导致可观测性差、责任边界模糊。为此，本文提出CHIEF框架，将混乱的执行轨迹转化为结构化的分层因果图；通过合成虚拟“仲裁者”引导回溯，高效剪枝搜索空间；最终采用渐进式因果筛查策略进行反事实归因，严格区分真实根因与传播症状。在Who&When基准上的实验表明，CHIEF在代理级和步骤级准确率上均优于八个强基线模型。消融实验进一步验证了各模块的关键作用。
原文摘要 · Abstract (English)
LLM-powered Multi-Agent Systems (MAS) have demonstrated remarkable capabilities in complex domains but suffer from inherent fragility and opaque failure mechanisms. Existing failure attribution methods, whether relying on direct prompting, costly replays, or supervised fine-tuning, typically treat execution logs as flat sequences. This linear perspective fails to disentangle the intricate causal links inherent to MAS, leading to weak observability and ambiguous responsibility boundaries. To address these challenges, we propose CHIEF, a novel framework that transforms chaotic trajectories into a structured hierarchical causal graph. It then employs hierarchical oracle-guided backtracking to efficiently prune the search space via sybthesized virtual oracles. Finally, it implements counterfactual attribution via a progressive causal screening strategy to rigorously distinguish true root causes from propagated symptoms. Experiments on Who&When benchmark show that CHIEF outperforms eight strong and state-of-the-art baselines on both agent- and step-level accuracy. Ablation studies further confirm the critical role of each proposed module.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。