A controlled synthetic-retrieval stress test shows that AI search agents differ greatly in handling untrustworthy, irrelevant, or false documents, and that HotpotQA is partly answerable from parametric knowledge alone.
Tracking the Limits of Knowledge Propagation: How LLM s Fail at Multi-Step Reasoning with Conflicting Knowledge
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DeepStress: Stress-Testing Deep Search Agents
A controlled synthetic-retrieval stress test shows that AI search agents differ greatly in handling untrustworthy, irrelevant, or false documents, and that HotpotQA is partly answerable from parametric knowledge alone.