用少量标注数据高效定制大模型，提升重症监护室诊疗预测与解释能力。
CU-ICU: Customizing Unsupervised Instruction-Finetuned Language Models for ICU Datasets via Text-to-Text Transfer Transformer
- 基于T5架构，结合少样本提示与参数选择性更新实现轻量微调。
- 脓毒症检测准确率最高提升15%，临床说明生成相关性增强20%。
- 仅更新不足1%参数，适合资源受限的医疗场景部署。
将大语言模型融入医疗等专业领域面临领域适配和标注数据稀缺的挑战。本文提出CU-ICU方法，利用Text-to-Text Transfer Transformer（T5）架构，对未监督指令微调的语言模型进行定制化，以适配重症监护室（ICU）数据集。该方法采用稀疏微调策略，结合少样本提示与选择性参数更新，实现低监督下的高效适应。在早期脓毒症检测、死亡率预测和临床记录生成等关键任务上的评估显示，相较于标准微调方法，CU-ICU在预测准确性和可解释性方面均有显著提升。尤其在最高效配置下，脓毒症检测准确率最高提升15%，生成临床相关解释的能力增强20%，同时更新的参数少于模型总量的1%。这些结果表明，CU-ICU是一种可扩展、低开销的解决方案，适用于现实世界中提供精准且可解释的临床决策支持。
原文摘要 · Abstract (English)
Integrating large language models into specialized domains like healthcare presents unique challenges, including domain adaptation and limited labeled data. We introduce CU-ICU, a method for customizing unsupervised instruction-finetuned language models for ICU datasets by leveraging the Text-to-Text Transfer Transformer (T5) architecture. CU-ICU employs a sparse fine-tuning approach that combines few-shot prompting with selective parameter updates, enabling efficient adaptation with minimal supervision. Our evaluation across critical ICU tasks--early sepsis detection, mortality prediction, and clinical note generation--demonstrates that CU-ICU consistently improves predictive accuracy and interpretability over standard fine-tuning methods. Notably, CU-ICU achieves up to a 15% increase in sepsis detection accuracy and a 20% enhancement in generating clinically relevant explanations while updating fewer than 1% of model parameters in its most efficient configuration. These results establish CU-ICU as a scalable, low-overhead solution for delivering accurate and interpretable clinical decision support in real-world ICU environments.
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。