用提示优化让大模型同时看病历和时间数据，预测更准还更灵活。
Integrating Text and Time-Series into (Large) Language Models to Predict Medical Outcomes
- 用DSPy优化提示，让大模型联合处理文本与电子病历时间序列
- 在临床分类任务上表现接近专用多模态系统
- 方法简单易改，适合多种医疗预测任务
大型语言模型（LLMs）在文本生成方面表现优异，但在涉及结构化数据（如时间序列）的临床分类任务中能力尚未充分探索。本文通过基于DSPy的提示优化，适配指令微调的大模型，使其能够联合处理临床笔记和结构化电子健康记录（EHR）输入。实验结果表明，该方法在性能上达到与专用多模态系统相当的水平，同时具备更低的复杂度和更强的任务泛化能力。
原文摘要 · Abstract (English)
Large language models (LLMs) excel at text generation, but their ability to handle clinical classification tasks involving structured data, such as time series, remains underexplored. In this work, we adapt instruction-tuned LLMs using DSPy-based prompt optimization to process clinical notes and structured EHR inputs jointly. Our results show that this approach achieves performance on par with specialized multimodal systems while requiring less complexity and offering greater adaptability across tasks.
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。