RuleAlign: Making Large Language Models Better Physicians with Diagnostic Rule Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Xiaohan, Yang, Xiaoyan, Zhu, Yuqi, Shen, Yue, Wang, Jian, Wei, Peng, Liang, Lei, Gu, Jinjie, Chen, Huajun, Zhang, Ningyu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916366480572416
author Wang, Xiaohan
Yang, Xiaoyan
Zhu, Yuqi
Shen, Yue
Wang, Jian
Wei, Peng
Liang, Lei
Gu, Jinjie
Chen, Huajun
Zhang, Ningyu
author_facet Wang, Xiaohan
Yang, Xiaoyan
Zhu, Yuqi
Shen, Yue
Wang, Jian
Wei, Peng
Liang, Lei
Gu, Jinjie
Chen, Huajun
Zhang, Ningyu
contents Large Language Models (LLMs) like GPT-4, MedPaLM-2, and Med-Gemini achieve performance competitively with human experts across various medical benchmarks. However, they still face challenges in making professional diagnoses akin to physicians, particularly in efficiently gathering patient information and reasoning the final diagnosis. To this end, we introduce the RuleAlign framework, designed to align LLMs with specific diagnostic rules. We develop a medical dialogue dataset comprising rule-based communications between patients and physicians and design an alignment learning approach through preference learning. Experimental results demonstrate the effectiveness of the proposed approach. We hope that our work can serve as an inspiration for exploring the potential of LLMs as AI physicians.
format Preprint
id arxiv_https___arxiv_org_abs_2408_12579
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RuleAlign: Making Large Language Models Better Physicians with Diagnostic Rule Alignment
Wang, Xiaohan
Yang, Xiaoyan
Zhu, Yuqi
Shen, Yue
Wang, Jian
Wei, Peng
Liang, Lei
Gu, Jinjie
Chen, Huajun
Zhang, Ningyu
Computation and Language
Artificial Intelligence
Human-Computer Interaction
Information Retrieval
Machine Learning
Large Language Models (LLMs) like GPT-4, MedPaLM-2, and Med-Gemini achieve performance competitively with human experts across various medical benchmarks. However, they still face challenges in making professional diagnoses akin to physicians, particularly in efficiently gathering patient information and reasoning the final diagnosis. To this end, we introduce the RuleAlign framework, designed to align LLMs with specific diagnostic rules. We develop a medical dialogue dataset comprising rule-based communications between patients and physicians and design an alignment learning approach through preference learning. Experimental results demonstrate the effectiveness of the proposed approach. We hope that our work can serve as an inspiration for exploring the potential of LLMs as AI physicians.
title RuleAlign: Making Large Language Models Better Physicians with Diagnostic Rule Alignment
topic Computation and Language
Artificial Intelligence
Human-Computer Interaction
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2408.12579