AI RESEARCH
Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories
arXiv CS.AI
•
ArXi:2606.02060v1 Announce Type: new Deep-research agents solve tasks through long trajectories of search, tool use, evidence inspection, and answer synthesis. Evaluation based on final answers shows whether an agent succeeds, but not which parts of the trajectory make the answer unreliable. We study span-level error localization for deep-research agents. We collect 2,790 real trajectories from two agent frameworks, three backbone models, and three benchmarks, convert raw logs into semantic spans, and annotate harmful error spans through LLM-assisted expert review.