Closing Reasoning Gaps in Clinical Agents with Differential Reasoning Learning

TL;DR

This paper introduces Differential Reasoning Learning (DRL), a framework that enhances clinical decision-making by identifying and correcting reasoning gaps through graph-based discrepancy analysis and retrieval-augmented reasoning, improving accuracy and fidelity.

Contribution

The paper presents a novel DRL framework that learns from reasoning discrepancies using graph-based analysis and retrieval, advancing clinical agent reasoning and reliability.

Findings

Improves answer accuracy in medical QA tasks.

Enhances reasoning fidelity with graph discrepancy analysis.

Clinician review confirms practical benefits.

Abstract

Clinical decision support requires not only correct answers but also clinically valid reasoning. We propose Differential Reasoning Learning (DRL), a framework that improves clinical agents by learning from reasoning discrepancies. From reference reasoning rationales (e.g., physician-authored clinical rationale, clinical guidelines, or outputs from more capable models) and the agent's free-form chain-of-thought (CoT), DRL extracts reasoning graphs as directed acyclic graphs (DAGs) and performs a clinically weighted graph edit distance (GED)-based discrepancy analysis. An LLM-as-a-judge aligns semantically equivalent nodes and diagnoses discrepancies between graphs. These graph-level discrepancy diagnostics are converted into natural-language instructions and stored in a Differential Reasoning Knowledge Base (DR-KB). At inference, we retrieve top-$k$ instructions via Retrieval-Augmented Generation (RAG) to augment the agent prompt and patch likely logic gaps. Evaluation on open medical question answering (QA) benchmarks and a Return Visit Admissions (RVA) prediction task from internal clinical data demonstrates gains over baselines, improving both final-answer accuracy and reasoning fidelity. Ablation studies confirm gains from infusing reference reasoning rationales and the top-$k$ retrieval strategy. Clinicians' review of the output provides further assurance of the approach. Together, results suggest that DRL supports more reliable clinical decision-making in complex reasoning scenarios and offers a practical mechanism for deployment under limited token budgets.