It is shown that evidence-gathering should be evaluated as a trajectory-level control problem, separately from answer-side reasoning, because larger hidden thinking budgets do not necessarily increase evidence inspection.
Abstract
Agentic retrieval-augmented generation (RAG) systems can fail before evidence-conditioned reasoning is tested: an agent may retrieve candidate snippets but finalize without inspecting them. We study this failure mode as a procedural property of the agent trajectory, decomposing wrong answers into pre-evidence discipline failures and post-gold-read failures using saved tool-call traces, retrieved evidence, read passages, and final answers. Across 12,000 paired trajectories on HotpotQA, 2WikiMultiHopQA, and MuSiQue, the two failure types are largely non-redundant: the both-trigger rate is in [11.2%, 13.1%] across regex and spaCy entity extractors. We then evaluate Read-Gate, a minimal runtime invariant requiring an agent to read after search and before finalization. Forced reading improves LLM-Acc by 14.9-19.9 points on trajectories that would otherwise skip reading and by 3.2-9.4 points on full minimal-reasoning cells. Additional diagnostics show that larger hidden thinking budgets do not necessarily increase evidence inspection. Together, these results indicate that evidence-gathering should be evaluated as a trajectory-level control problem, separately from answer-side reasoning.
This work introduces SLEUTH, which makes an agent's investigative state explicit and actionable through a structured epistemic working memory, and shows that how an agent organizes its reasoning, not raw model capability, is the active ingredient for scaling multi-hop reasoning.
Agentic retrieval-augmented generation (RAG) interleaves retrieval, reasoning, and answer generation across multiple hops. A retrieval error at hop 1 can surface only as a wrong answer at hop 3, while later retrieval can also repair the trajectory. This paper introduces AgenticRAG-FP, an interventional benchmark for causal failure attribution in agentic RAG. The benchmark injects a certified fault at a specified hop, re-executes the downstream trajectory, and evaluates diagnosers against the known intervention. Its central question is whether a post-hoc trace still identifies the injected hop after the suffix changes. In the completed strict dense Claude Haiku 4.5 sweep on 80 three-hop MuSiQue questions, coverage-based diagnosis is 0.91 at hop 1 and 0.00 at hops 2 and 3 (n=43,36,21 failed trajectories). A smaller content-corruption study changes an answer-bearing or bridge fact in topically intact evidence. At depth 2, where 18 failed cases remain after filtering, coverage-based diagnosis is 0.00 and a frozen-hop counterfactual probe is 0.67 in an exploratory pooled comparison. Depth-3 content estimates are descriptive only because they contain three failed cases. These results make propagation depth an explicit evaluation axis for diagnosing agentic RAG failures while distinguishing broad evidence of post-hoc signal loss from small-sample method comparisons.
This work introduces HopRefusalBench, the first controlled benchmark of refusal within multi-hop search, and proposes a final-outcome taxonomy spanning target-aware refusal, pseudo-refusal, hallucinated completion, and search-budget exhaustion, together with source-aware trajectory metrics for post-trigger continuation and token waste.
Jianan Xie, Xin Sun, Zhongqi Chen et al.· 0 citations
SymStep: an LLM makes one atomic claim at a time (DEDUCE: Alice, pet, Cat), then a lightweight constraint propagator checks the claim for consistency with prior accepted deductions, rejects contradictions, and cascades implied facts automatically.
Reason Popper-ly, a neurosymbolic framework that uses inductive logic programming (ILP) to learn relation composition rules from reasoning traces and deploys them as an online verifier for step-level correction, consistently improves terminal accuracy over standard CoT.
This work introduces Harness-IF, which scores operational rules one at a time from execution evidence: 60 realistic multi-turn coding items drawn from a 642-rule library, 256 rules receiving verdicts, placed on the five configurable surfaces a deployed agent reads.
Zining Huang, Haoran Que, Hongxia Zeng et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
