Taking the PULSE of Medical Reasoning: Step-Level Retrieval Timing for Knowledge-Grounded Inference

arXiv:2610.20292v1 Announce Type: new Abstract: Reliable medical reasoning requires LLMs to dynamically access external biomedical evidence to improve factual trustworthiness and reasoning accuracy. However, indiscriminate retrieval can introduce noise and latency, raising a critical retrieval-timing problem: at each reasoning step, the system must decide whether the model's internal knowledge is reliable or external evidence is needed. We propose PUlSE, a lightweight probing method that uses the model's internal representations as intrinsic signals for step-level retrieval control. Our posterior trajectory analysis reveals that erroneous steps exhibit substantially higher inter-layer volatility than correct ones, suggesting unstable internal knowledge. Based on this finding, PUlSE trains a compact probe on inter-layer geometric features to detect factual errors upon step completion, and uses it as an adaptive trigger for retrieval and step regeneration. Across four medical benchmarks, PUlSE-timed retrieval matches exhaustive per-step retrieval in accuracy while reducing retrieval calls by 72%, and can transfer to an unseen legal domain, indicating potential domain-agnostic generality. Overall, our study provides an initial step toward shifting retrieval-trigger design from surface-level confidence estimates to internal activation dynamics for adaptive access.

Related

• MIRAGE: Defending Long-Form RAG Against Misinformation Pollution
• VeriLLMed: Interactive Visual Debugging of Medical Large Language Models with Knowledge Graphs
• Hierarchical Retrieval with Out-Of-Vocabulary Queries: A Case Study on SNOMED CT
• Where Does Retrieval-Based Open-Ended Evaluation Fail? Automatic Taxonomy Induction from Long-Form Medical Answer Factuality Verification

Source: arXiv cs.CL | 2026-10-15