Improving Generalizability of Extracting Social Determinants of Health Using Large Language Models through Prompt-tuning
arxiv(2024)
摘要
The progress in natural language processing (NLP) using large language models
(LLMs) has greatly improved patient information extraction from clinical
narratives. However, most methods based on the fine-tuning strategy have
limited transfer learning ability for cross-domain applications. This study
proposed a novel approach that employs a soft prompt-based learning
architecture, which introduces trainable prompts to guide LLMs toward desired
outputs. We examined two types of LLM architectures, including encoder-only
GatorTron and decoder-only GatorTronGPT, and evaluated their performance for
the extraction of social determinants of health (SDoH) using a
cross-institution dataset from the 2022 n2c2 challenge and a cross-disease
dataset from the University of Florida (UF) Health. The results show that
decoder-only LLMs with prompt tuning achieved better performance in
cross-domain applications. GatorTronGPT achieved the best F1 scores for both
datasets, outperforming traditional fine-tuned GatorTron by 8.9
cross-institution setting, and 5.5
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要