Why Your Deep Research Agent Fails? On Hallucination Evaluation in Full Research Trajectory 文章

arXiv:2601.22984v2 Announce Type: replace Abstract: Diagnosing failure patterns in Deep Research Agents (DRAs) remains a critical challenge. Existing benchmarks predominantly rely on end-to-end evaluation, obscuring intermediate hallucinations that accumulate throughout the research trajectory. To bridge this gap, we propose a shift from outcome-based to processaware evaluation by auditing hallucinations in the full plan-search-summarize trajectory. We introduce the PING Taxonomy, which categorizes DRA hallucinations into four complementary types: Propagation, Intent, Noiseinduced, and Grounding. We further instantiate this taxonomy into a fine-grained evaluation framework that decomposes trajectories into atomic actions, claims, and sub-queries for rigorous verification. Leveraging this framework to isolate 100 distinctively hallucinationprone tasks including adversarial scenarios, we curate DeepHalluBench.