DeFA: Dependency-Guided Failure Attribution for LLM Agents
DeFA is a new framework that helps identify the root causes of errors in LLM agent executions by constructing an event dependency graph and a failure propagation graph.
01
Evidence
- AarXiv cs.AIPrimary source2026-10-02 12:00
> Abstract: Errors in LLM agent executions and their visible consequences can be separated by many steps, making decisive-error localization a matter of understanding both step content and step dependencies.
- AarXiv cs.AIPrimary source2026-10-02 12:00
Abstract: Errors in LLM agent executions and their visible consequences can be separated by many steps, making decisive-error localization a matter of understanding both step content and step dependencies. We introduce DeFA, a dependency-guided framework for agent failure attribution. DeFA first combines protocol relations and semantic dependencies into an event dependency graph spanning the traj…