定位智能体系统故障根源，精准找出最小修复组合。
Localizing Emergent Failures in Agentic AI: Recovering Minimal Repair Families via Counterfactual Replay
- 从执行依赖图中提取故障相关事件，构造候选修复对
- 在90个案例中实现100%修复家族匹配，回放调用减少55.1%
- 适合需精准定位多智能体系统故障的开发者和研究者
智能体系统中的失败可能源于多个大语言模型智能体之间消息交互所致。单一归因无法区分联合必要修复与独立修复方案。本文提出最小修复家族恢复（MRFR）：在指定规模范围内，恢复所有包含最小事件集的反事实重播可恢复任务成功的集合。我们提出图约束联合重播（GCJR），从执行依赖图中切分故障相关事件，构建图可行的单事件与成对候选修复，并通过配对干净对照重播验证。在120个有向无环图（DAG）控制基准的90个案中，GCJR实现1.000的家族精确匹配，平均重播调用从56.3降至25.3（减少55.1%）；在含4个智能体的24个案例的试点中，再次达成1.000精确匹配，平均模型调用从21.0降至10.0（减少52.4%）；单事件重播会遗漏联合必要修复。
原文摘要 · Abstract (English)
Failures in agentic AI systems can arise from interactions among messages exchanged by multiple large language model (LLM) agents. Pointwise attribution cannot distinguish a jointly necessary repair from alternative singleton repairs. We formulate Minimal Repair Family Recovery (MRFR): recovering all inclusion-minimal event sets whose counterfactual replay restores task success within a declared size bound. We propose Graph-Constrained Joint Replay (GCJR), which slices failure-relevant events from an execution dependency graph, constructs graph-feasible singleton and pair candidates, and verifies them by replay with paired clean counterparts. For fixed replay outcomes, GCJR is exact within its declared graph domain. On 90 in-scope cases from a 120-DAG controlled benchmark, GCJR achieves 1.000 Family Exact Match while reducing mean replay calls from 56.3 to 25.3 (55.1%) relative to exhaustive search. On a 24-case, four-agent LLM pilot, it again achieves 1.000 Family Exact Match and reduces mean model calls from 21.0 to 10.0 (52.4%); single-event replay misses jointly necessary repairs.
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。