An Empirical Evaluation of Prompting Strategies for Large Language Models in Zero-Shot Clinical Natural Language Processing: Algorithm Development and Validation Study

• Published in: JMIR medical informatics Vol. 12; p. e55318
• Main Authors: Sivarajkumar, Sonish, Kelley, Mark, Samolyk-Mazzanti, Alyssa, Visweswaran, Shyam, Wang, Yanshan
• Format: Journal Article
• Language: English
• Published: Canada JMIR Publications 08.04.2024
• Subjects: Annotations; Classification; Datasets; Drug dosages; Engineering; Evidence-based medicine; Language; Large language models; Medical research; Natural language processing; Original Paper; Performance evaluation; Text categorization; GPT; ensemble; language; LLM; clinical information; large language model; clinical data; in-context learning; evaluation; NLP; few shot; prompting; LLaMA-2; prompts; prompt engineering; models; natural language processing; LLMs; heuristic; language model; extraction; machine learning; zero-shot; BARD; prompt; Gemini

Large language models (LLMs) have shown remarkable capabilities in natural language processing (NLP), especially in domains where labeled data are scarce or expensive, such as the clinical domain. However, to unlock the clinical knowledge hidden in these LLMs, we need to design effective prompts tha...