← Back to all news
arXiv AI By Yarden Bakish, Amir Dudai, Roy Ganz, Oren Nuriel, Elad Ben Avraham, Mor Shpigel Nacson, Ron Litman

Adaptive Influence Graphs for Failure Attribution in Multi-Agent Systems

Read the original on arXiv AI →

The paper introduces Adaptive Influence Graphs (AIGs), a two‑stage framework that first converts a failed trace into a structured graph and then navigates it to pinpoint the critical error in multi‑agent large language model systems. Experiments across multiple models demonstrate that richer trace representations and adaptive graph construction improve failure attribution, with AIGs achieving state‑of‑the‑art results on the Who&When benchmark. The study shows that both the diagnosing model and the way traces are represented and explored are crucial for accurate failure attribution.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Sep 2

EDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systems

EDGE is a framework that attributes multiple related errors in multi-agent large language model systems by constructing an error dependency graph from observed error events. It validates a reliable causal subset through counterfactual rollout and uses this inference graph to guide a two-stage LLM-as-judge detector for error attribution. Experiments on TRAIL and MAST demonstrate that EDGE improves category-level multi-error attribution across most models and settings, and that the graph aids explanation and repair analysis.

By Jun Hou, Priya Pitre, Yi Fang, Xuan Wang
llmsagents
More like this →
arXiv AI
Sep 7

DCFA: Dual-view Causal-inspired Attribution for Failure Reasoning in LLM-based Multi-agent Systems

The paper introduces DCFA, a training‑free framework for attributing failures in large language model‑based multi‑agent systems. DCFA uses a global module to build causal‑inspired dependency graphs from system traces, pinpointing the earliest decisive error, and a local module that refines this attribution through counterfactual reasoning. Experiments on the Who&When benchmark across six LLMs demonstrate that DCFA improves step‑level accuracy by up to 8.27% over existing baselines.

By Zehao Wang, Lanjun Wang, Shilong Jin, Junjie Chen, Yanghua Xiao
llmsagentsbenchmarks
More like this →