← 研踪 PaperTrailarXiv:2512.03838 · v2
Guideline-Based Medical Reasoning in LLMs
首次提交 2025-12-03 22:39
版本更新 2026-11-16 21:51
雷达观测 2026-11-17 17:11
摘要
Large Language Models (LLMs) excel at medical question-answering, yet progress toward medical reasoning models is limited by the lack of scalable, trustworthy reasoning supervision. We propose tapping into the so far underused resource of medical consensus guidelines. These map numerical clinical measurements to disease scores and can optionally be curated for site-specific exceptions. Automatically generating gold-standard reasoning chains from consensus rules enables both large-scale supervised fine-tuning and automatic evaluation. Applying SOFA, SAPS-II, and Sepsis-3 guidelines to MIMIC-III, we show that small fine-tuned models significantly outperform larger one-shot prompted models in derivation correctness and F1 score, achieving near-perfect deductive inference and reliable exception handling. Trustworthy step-by-step reasoning does not reliably emerge from one-shot prompting, even when models receive the consensus definition and an example. Performance across all models declines when forecasting future clinical variables, indicating that, on the evaluated structured ICU tasks, the central challenge is not rule application, but the orthogonal problem of generalizing into the future for prognostic reasoning.
作者备注:v2: camera-ready version, accepted at ML4H 2026; title changed
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-16 21:51 · 最新观测版本
Guideline-Based Medical Reasoning in LLMs
Large Language Models (LLMs) excel at medical question-answering, yet progress toward medical reasoning models is limited by the lack of scalable, trustworthy reasoning supervision. We propose tapping into the so far underused resource of medical consensus guidelines. These map numerical clinical measurements to disease scores and can optionally be curated for site-specific exceptions. Automatically generating gold-standard reasoning chains from consensus rules enables both large-scale supervised fine-tuning and automatic evaluation. Applying SOFA, SAPS-II, and Sepsis-3 guidelines to MIMIC-III, we show that small fine-tuned models significantly outperform larger one-shot prompted models in derivation correctness and F1 score, achieving near-perfect deductive inference and reliable exception handling. Trustworthy step-by-step reasoning does not reliably emerge from one-shot prompting, even when models receive the consensus definition and an example. Performance across all models declines when forecasting future clinical variables, indicating that, on the evaluated structured ICU tasks, the central challenge is not rule application, but the orthogonal problem of generalizing into the future for prognostic reasoning.
查看此版本解读 →查看此版本原文 ↗