[Submitted on 15 Jul 2026 (v1), last revised 12 Nov 2026 (this version, v2)]
Abstract:While search agents demonstrate impressive capabilities in multi-step question answering, their robustness to poor-quality evidence remains under-explored. Such evidence is rare in benchmark corpora, and resulting failures are hard to reproduce. We propose DeepStress, a dynamic stress-testing framework that subjects agents to degraded evidence at controlled rates. It replaces the retrieval module with a synthetic environment where documents are generated dynamically for each query, under independent control of their trustworthiness, relevance, and factuality. To characterize behavior under stress, we go beyond binary accuracy with a five-way classification of final-answer outcomes and two new reliability-aware metrics, RAS and CoPRAS, which credit cautious behavior and account for computational cost. Evaluating 12 systems on questions from HotpotQA and BrowseComp-Plus, we find that several open-weight agents answer confidently when sources contain obvious credibility red flags, and that many HotpotQA questions are answered from parametric knowledge even when all returned documents are factually wrong.
Submission history
From: Ismaël Rousseau [view email]
[v1] Wed, 15 Jul 2026 14:59:29 UTC (1,247 KB)
[v2] Thu, 12 Nov 2026 13:18:48 UTC (1,271 KB)
