用对话式推理让大模型生成可解释的连续风险评分。
TRIAGE: Dialectical Reasoning for Explainable Risk Prediction on Irregularly Sampled Medical Time Series with LLMs
- 让大模型对比多种临床结局，生成辩证性推理
- 风险评分连续且校准度提升81%，AUPRC平均提高3.3%
- 适合需要可解释医疗预警的临床研究与系统开发
基于电子病历的临床早期预警系统处理的是不规则采样的医疗时间序列（ISMTS），需提供校准的风险评分和医生可验证的可解释理由。尽管大语言模型（LLMs）被用于此任务，但它们常将分级风险压缩为过度自信的二元预测，导致风险极化，损害校准性和跨患者可比性。为此，我们提出TRIAGE框架，训练大模型通过提取特定结局的理由，进行竞争性临床结局的辩证推理。该方法缓解了风险极化，使单一模型能输出基于明确临床推理的连续风险评分。在三个ISMTS基准上评估，TRIAGE相比基线平均提升3.3% AUPRC，校准误差降低81%。以大模型为评判者评估显示，我们的推理质量比基线后验解释高出20%。代码已开源：https://github.com/HyeongWon-Jang/TRIAGE。
原文摘要 · Abstract (English)
Clinical early warning systems built on electronic health records, in which clinical observations are recorded as irregularly sampled medical time series (ISMTS), must deliver both calibrated risk scores for patient triage and interpretable rationales that clinicians can verify. Large Language Models (LLMs) have been explored for this task, yet they collapse graded clinical risk into overconfident binary predictions. This risk polarization undermines both calibration and cross-patient comparability. To address this, we propose TRIAGE, a framework that trains an LLM to generate dialectical reasoning over competing clinical outcomes by eliciting outcome-specific rationales. This dialectical formulation mitigates risk polarization, enabling a single LLM to yield continuous risk scores grounded in explicit clinical reasoning. Evaluated on three ISMTS benchmarks, TRIAGE achieves an average AUPRC improvement of 3.3% and reduces calibration error by 81% compared to the competitive baselines. An LLM-as-a-judge assessment further shows that our rationales surpass post-hoc explanations from the baseline by 20% in clinical reasoning quality. The source code is available at https://github.com/HyeongWon-Jang/TRIAGE .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。