From Flat Logs to Causal Graphs: Hierarchical Failure Attribution for LLM-based Multi-Agent Systems

2026-02-27

AI Summary

Existing approaches treat execution logs of large language model (LLM)-driven multi-agent systems as flat sequences, failing to uncover causal relationships and assign responsibility within complex interactions. This work introduces a hierarchical causal graph to structurally model execution trajectories and proposes a counterfactual oracle-guided backtracking mechanism coupled with a progressive counterfactual filtering strategy to enable fine-grained fault attribution. Evaluated on the Who&When benchmark, the proposed method significantly outperforms eight state-of-the-art baselines in both agent-level and step-level attribution accuracy. Ablation studies further confirm the effectiveness of each component in the proposed framework.

Technology Category

• Multiagent Systems: Adversarial Agents
• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

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.

Problem

• LLM-based Multi-Agent Systems
• failure attribution
• causal reasoning
• execution logs
• observability

Innovation

• causal graph
• hierarchical attribution
• counterfactual reasoning
• multi-agent systems
• LLM-based systems