EpilepsyLLM: Domain-Specific Large Language Model Fine-tuned with Epilepsy Medical Knowledge

TL;DR

General LLMs struggle with domain-specific medical knowledge in languages other than English. The approach uses fine-tuning of pre-trained bases such as LLaMA and LLM-JP with a Japanese epilepsy knowledge dataset presented as instruction-following demonstrations. The main finding is that EpilepsyLLM achieves superior epilepsy domain performance, with the best results from LLM-JP 1.3B after fine-tuning, and that language alignment and data translation can yield additional gains. This work demonstrates the value of targeted, language-specific domain adaptation for non-English medical NLP, offering more reliable epilepsy information in Japanese while guiding future domain-focused fine-tuning efforts.

Abstract

With large training datasets and massive amounts of computing sources, large language models (LLMs) achieve remarkable performance in comprehensive and generative ability. Based on those powerful LLMs, the model fine-tuned with domain-specific datasets posseses more specialized knowledge and thus is more practical like medical LLMs. However, the existing fine-tuned medical LLMs are limited to general medical knowledge with English language. For disease-specific problems, the model's response is inaccurate and sometimes even completely irrelevant, especially when using a language other than English. In this work, we focus on the particular disease of Epilepsy with Japanese language and introduce a customized LLM termed as EpilepsyLLM. Our model is trained from the pre-trained LLM by fine-tuning technique using datasets from the epilepsy domain. The datasets contain knowledge of basic information about disease, common treatment methods and drugs, and important notes in life and work. The experimental results demonstrate that EpilepsyLLM can provide more reliable and specialized medical knowledge responses.