MedGuideX: Internalizing Decision Logic from Executable Guidelines into Large Language Models for Clinical Reasoning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Shen, Yuhao, Cao, Lang, Du, Simo, Wang, Yuqing, Zhou, Juexiao, Peng, Hao, Guo, Yue
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918523590148096
author Shen, Yuhao
Cao, Lang
Du, Simo
Wang, Yuqing
Zhou, Juexiao
Peng, Hao
Guo, Yue
author_facet Shen, Yuhao
Cao, Lang
Du, Simo
Wang, Yuqing
Zhou, Juexiao
Peng, Hao
Guo, Yue
contents Clinical practice guidelines (CPGs) encode evidence-based decision logic that clinicians apply by evaluating patient variables, conditional criteria, and recommendation rules. However, existing methods often use CPGs as free-text training data or retrieval sources, underutilizing their procedural decision structure. To better exploit this structure, we introduce a guideline-derived training pipeline that transforms CPG recommendations into executable clinical decision logic and uses it to generate factual and counterfactual question-answering data. Theses data teach models both guideline-supported decisions and how decisions change under different patient conditions. Post-training a medical LLM on the generated data yields MedGuideX. Across four clinical reasoning benchmarks, MedGuideX achieves a 10.28% relative improvement in average accuracy. Physician evaluation further shows that MedGuideX better recovers clinician authored reasoning steps and produces physician-preferred rationales in faithfulness, validity, completeness, and clarity. Overall, our results show that executable decision logic from CPGs can be transformed into scalable supervision for building reliable medical LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2605_26567
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MedGuideX: Internalizing Decision Logic from Executable Guidelines into Large Language Models for Clinical Reasoning
Shen, Yuhao
Cao, Lang
Du, Simo
Wang, Yuqing
Zhou, Juexiao
Peng, Hao
Guo, Yue
Artificial Intelligence
Clinical practice guidelines (CPGs) encode evidence-based decision logic that clinicians apply by evaluating patient variables, conditional criteria, and recommendation rules. However, existing methods often use CPGs as free-text training data or retrieval sources, underutilizing their procedural decision structure. To better exploit this structure, we introduce a guideline-derived training pipeline that transforms CPG recommendations into executable clinical decision logic and uses it to generate factual and counterfactual question-answering data. Theses data teach models both guideline-supported decisions and how decisions change under different patient conditions. Post-training a medical LLM on the generated data yields MedGuideX. Across four clinical reasoning benchmarks, MedGuideX achieves a 10.28% relative improvement in average accuracy. Physician evaluation further shows that MedGuideX better recovers clinician authored reasoning steps and produces physician-preferred rationales in faithfulness, validity, completeness, and clarity. Overall, our results show that executable decision logic from CPGs can be transformed into scalable supervision for building reliable medical LLMs.
title MedGuideX: Internalizing Decision Logic from Executable Guidelines into Large Language Models for Clinical Reasoning
topic Artificial Intelligence
url https://arxiv.org/abs/2605.26567