Right Diagnoses, Decorative Reasoning:A Perturbation Audit of Medical Chain-of-Thought

• URL: http://arxiv.org/abs/2608.24790v1
• Date: Tue, 25 Aug 2026 16:37:39 GMT
• Title: Right Diagnoses, Decorative Reasoning:A Perturbation Audit of Medical Chain-of-Thought
• Abstract summary: Clinicians read chain-of-thought (CoT) rationales as evidence of medical reasoning, but whether the visible chain plays that role is rarely tested.<n>Three independent tests converge: the Chain-Decoupling Rate (CDR; chain does not register the edit and the answer does not flip) is 72.9% panel-wide on clinically meaningful destructive edits, chain corruption leaves accuracy unchanged, and removing CoT prompting does not reduce accuracy.
• Score: 20.946393640258997
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Clinicians read chain-of-thought (CoT) rationales as evidence of medical reasoning, but whether the visible chain plays that role is rarely tested. General-domain CoT-faithfulness probes ignore clinical cost, and medical LLM evaluations treat the chain as a black box. We close this gap with a medical perturbation audit: a 30-operator battery edits both the chain and the question with clinically motivated operators (severity reversal, negation flip, demographic swap, evidence ablation), paired with a chain-update times answer-flip joint analysis that classifies each model by its failure mode. Applied to 14 LLMs on four medical QA benchmarks, three independent tests converge: the Chain-Decoupling Rate (CDR; chain does not register the edit and the answer does not flip) is 72.9% panel-wide on clinically meaningful destructive edits, chain corruption leaves accuracy unchanged, and removing CoT prompting does not reduce accuracy. Two board-certified clinicians re-annotate N=197 perturbed questions; 98.5% leave the gold defensible. The pattern holds across medical and reasoning fine-tuning and scale; on the closed-source tier, where the chain text is unavailable, the answer-side signals are consistent with the same decoupling. Our framework and CDR provide a reusable yardstick for auditing whether medical CoT is faithful or merely documentation.

Related papers

• MedSNIP: Building and Benchmarking Snippet-Level Granularity for Medical Fact Verification [50.13873960887466]
  We introduce MedSNIP-Bench, a benchmark for snippet-level medical fact verification.<n>We also introduce MedSNIP, an automatic snippet-generation pipeline.<n>Across MedSNIP-Bench, HealthFC, and MedHallu, snippet-level verification preserves or improves false-class F1.
  arXiv  Detail & Related papers  (2026-09-11T14:08:44Z)
• When Patients Cut In: Extending Clinical Conversational AI Safety to Interruptions [1.759008116536278]
  Clinical voice agents are now deployed in routine care, where real patients do not wait their turn: they interrupt.<n>Clinical conversational-AI benchmarks almost universally assume patients wait for the agent to finish, missing interruption-induced loss of required content.<n>We present a transcript-based evaluation of interruption recovery, adapting conversation-analytic overlap categories into three operational types.
  arXiv  Detail & Related papers  (2026-08-29T12:54:50Z)
• Task-Conditional Faithfulness Auditing of Multimodal LLMs for Grid Diagnosis [5.427346259545067]
  Multimodal large language models (LLMs) can combine topology, measurements, and incident text for grid diagnosis.<n>This letter proposes a general framework in order to conduct task-conditional faithfulness audit.
  arXiv  Detail & Related papers  (2026-07-27T15:17:15Z)
• Beyond Scalar Scores: Exploring LLM-based Metrics for Clinical Significance Evaluation in Radiology Reports [49.5225801722164]
  Existing metrics obscure this requirement by reducing report quality to a medically ungrounded scalar.<n>We study this boundary using ReEvalMed benchmark as testbed and evaluate metric-level clinical significance.
  arXiv  Detail & Related papers  (2026-06-17T08:10:30Z)
• Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications [17.69274632062373]
  Open-ended conversational agents remain prone to two critical failure modes: premature diagnostic handoff and silent clinical hallucinations.<n>We propose a multi-agent framework that addresses both issues by replacing LLM-as-a-judge'' routing with deterministic orchestration constraints.<n>We evaluate the system using simulated patient agents powered by the llama-3.1-70b-instruct model on 150 test cases.
  arXiv  Detail & Related papers  (2026-06-16T15:39:19Z)
• What Makes a Medical Checker Trainable? Diagnosing Signal Collapse and Reward Hacking in Checker-Guided RAG for Biomedical QA [6.163604976615509]
  Medical RAG needs evidence-grounded claims, so plugging a claim-level NLI checker into retrieval-augmented RL is intuitive.<n>We compare four NLI checker back-ends as process rewards inside a GRPO-trained medical RAG agent.
  arXiv  Detail & Related papers  (2026-05-25T16:06:06Z)
• CXR-ContraBench: Benchmarking Negated-Option Attraction in Medical VLMs [20.96410413299322]
  CXR-ContraBench is a diagnostic benchmark spanning internal ReXVQA slices and external OpenI and CheXpert protocols.<n>We study this failure as negated-option attraction, where a model is drawn to a negated answer option even when it conflicts with both the visual evidence and the question.
  arXiv  Detail & Related papers  (2026-05-07T07:46:17Z)
• FACT-E: Causality-Inspired Evaluation for Trustworthy Chain-of-Thought Reasoning [49.65751420291115]
  Chain-of-Thought (CoT) prompting has improved LLM reasoning, but models often generate explanations that appear coherent while containing unfaithful intermediate steps.<n>We propose FACT-E, a causality-inspired framework for evaluating CoT quality.<n>We show that FACT-E improves reasoning-trajectory selection and yields stronger in-context learnings.
  arXiv  Detail & Related papers  (2026-04-12T15:35:08Z)
• Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction [72.89352076103889]
  Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn.<n>We introduce MINT, a high-fidelity, multi-turn medical diagnosis benchmark comprising 1,035 cases.<n>We uncover three persistent behavioral patterns that significantly impact diagnostic decisions.
  arXiv  Detail & Related papers  (2026-04-06T00:23:10Z)
• From Conflict to Consensus: Boosting Medical Reasoning via Multi-Round Agentic RAG [65.0203623486525]
  Large Language Models (LLMs) exhibit high reasoning capacity in medical question-answering.<n>Their tendency to produce hallucinations and outdated knowledge poses critical risks in healthcare fields.<n>Existing methods rely on noisy token-level signals and lack the multi-round refinement required for complex reasoning.
  arXiv  Detail & Related papers  (2026-02-06T08:25:30Z)
• M3CoTBench: Benchmark Chain-of-Thought of MLLMs in Medical Image Understanding [66.78251988482222]
  Chain-of-Thought (CoT) reasoning has proven effective in enhancing large language models by encouraging step-by-step intermediate reasoning.<n>Current benchmarks for medical image understanding generally focus on the final answer while ignoring the reasoning path.<n>M3CoTBench aims to foster the development of transparent, trustworthy, and diagnostically accurate AI systems for healthcare.
  arXiv  Detail & Related papers  (2026-01-13T17:42:27Z)
• MEDEQUALQA: Evaluating Biases in LLMs with Counterfactual Reasoning [7.167933033102407]
  We introduce MEDEQUALQA, a counterfactual benchmark that perturbs only patient pronouns while holding critical symptoms and conditions constant.<n>We evaluate a GPT-4.1 model and compute Semantic Textual Similarity (STS) between reasoning traces to measure stability across pronoun variants.<n>Our results show overall high similarity (mean STS >0.80), but reveal consistent localized divergences in cited risk factors, guideline anchors, and differential ordering.
  arXiv  Detail & Related papers  (2025-10-09T22:12:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.