先检测异常再归因，提升多智能体系统故障定位准确率
Detect Before You Attribute: Cascade Failure Attribution for Multi-Agent Systems
- 先用双视角表示和树形LSTM检测异常执行轨迹
- 在六种基线方法上提升整体与步骤级归因准确率
- 适合需要高可靠性的复杂任务多智能体系统
基于大语言模型（LLM）的智能体在多步推理中展现强大潜力，但仍易发生执行失败。准确的故障归因对提升智能体可靠性至关重要。现有拓扑与谱方法虽利用轨迹结构，却常忽略细粒度语义；而基于LLM的方法虽捕捉语义线索，但长轨迹下存在上下文退化问题。为此，我们提出DUOTRACE，一种即插即用的检测过滤器，采用‘先检测后归因’范式：先识别异常执行，再向下游归因方法提供聚焦的轨迹证据。为实现高效的基于变分自编码器（VAE）的异常检测，DUOTRACE融合双视角语义-结构节点表示、树形LSTM轨迹编码器，以及前缀链与基于LLM的数据增强，以应对异构节点、层级执行结构及有限失败数据的问题。在六个基于LLM的归因基线上实验表明，DUOTRACE使智能体级和步骤级归因准确率分别提升8.7%和7.0%。
原文摘要 · Abstract (English)
Large language model (LLM)-based agents have shown strong potential in solving complex tasks through multi-step reasoning, yet they remain vulnerable to execution failures. Accurate failure attribution is therefore critical for improving agent reliability. Existing topology- and spectrum-based methods exploit trajectory structures but often overlook fine-grained semantics, while LLM-based attribution methods capture semantic cues but suffer from long-context degradation over lengthy trajectories. To address these challenges, we propose DUOTRACE, a plug-and-play detection filter for LLM-based failure attribution. DUOTRACE follows a detect-before-attribute paradigm: it first detects anomalous executions and then supplies focused trajectory evidence to downstream LLM-based attribution methods. For effective VAE-based anomaly detection on agent trajectories, DUOTRACE integrates dual-view semantic-structural node representations, a Tree-LSTM-based trajectory encoder, and prefix-chain- and LLM-based data augmentation to handle heterogeneous nodes, hierarchical execution structures, and limited failure data. Experiments with six LLM-based attribution baselines show that DUOTRACE improves agent-level and step-level attribution accuracy by 8.7% and 7.0%, respectively.
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。