Beyond Fine-tuning: Unleashing the Potential of Continuous Pretraining for Clinical LLMs

TL;DR

This paper explores various techniques like continuous pretraining, instruct fine-tuning, NEFTune, and prompt engineering to enhance large language models for clinical applications, highlighting their individual and combined impacts.

Contribution

It systematically evaluates the effectiveness of multiple adaptation techniques for clinical LLMs, introducing NEFTune and analyzing their interplay.

Findings

Continuous pretraining beyond 250B tokens provides marginal gains alone.

NEFTune improves generation quality and overall performance.

Prompt engineering further boosts clinical task performance.

Abstract

Large Language Models (LLMs) have demonstrated significant potential in transforming clinical applications. In this study, we investigate the efficacy of four techniques in adapting LLMs for clinical use-cases: continuous pretraining, instruct fine-tuning, NEFTune, and prompt engineering. We employ these methods on Mistral 7B and Mixtral 8x7B models, leveraging a large-scale clinical pretraining dataset of 50 billion tokens and an instruct fine-tuning dataset of 500 million tokens. Our evaluation across various clinical tasks reveals the impact of each technique. While continuous pretraining beyond 250 billion tokens yields marginal improvements on its own, it establishes a strong foundation for instruct fine-tuning. Notably, NEFTune, designed primarily to enhance generation quality, surprisingly demonstrates additional gains on our benchmark. Complex prompt engineering methods further enhance performance. These findings show the importance of tailoring fine-tuning strategies and exploring innovative techniques to optimize LLM performance in the clinical domain.