DoctorGLM: Fine-tuning your Chinese Doctor is not a Herculean Task

Honglin Xiong, Sheng Wang, Yitao Zhu, Zihao Zhao, Yuxiao Liu +3 more
4/3/2023
cs.CL

Official code: xionghonglin/DoctorGLM (833 stars, last commit 2023-10-19).

Abstract

The recent progress of large language models (LLMs), including ChatGPT and GPT-4, in comprehending and responding to human instructions has been remarkable. Nevertheless, these models typically perform better in English and have not been explicitly trained for the medical domain, resulting in suboptimal precision in diagnoses, drug recommendations, and other medical advice. Additionally, training and deploying a dialogue model is still believed to be impossible for hospitals, hindering the promotion of LLMs. To tackle these challenges, we have collected databases of medical dialogues in Chinese with ChatGPT's help and adopted several techniques to train an easy-deploy LLM. Remarkably, we were able to fine-tune the ChatGLM-6B on a single A100 80G in 13 hours, which means having a healthcare-purpose LLM can be very affordable. DoctorGLM is currently an early-stage engineering attempt and contain various mistakes. We are sharing it with the broader community to invite feedback and suggestions to improve its healthcare-focused capabilities: https://github.com/xionghonglin/DoctorGLM.

View on arXivView PDF

Code Implementations(1)

xionghonglin/DoctorGLMOfficial100%

基于ChatGLM-6B的中文问诊模型

83388HTML, PythonMar 29, 20233 years ago
chatglm-6b

Similar Papers

Qilin-Med: Multi-stage Knowledge Injection Advanced Medical Large Language Model

89% match1 repo162023

Continuous Training and Fine-tuning for Domain-Specific Language Models in Medical Question Answering

89% match2023

MedChatZH: a Better Medical Adviser Learns from Better Instructions

89% match1 repo1102023

JMedLoRA:Medical Domain Adaptation on Japanese Large Language Models using Instruction-tuning

89% match2023

IvyGPT: InteractiVe Chinese pathwaY language model in medical domain

88% match1 repo572023

ChatDoctor: A Medical Chat Model Fine-Tuned on a Large Language Model Meta-AI (LLaMA) Using Medical Domain Knowledge

88% match1 repo3.6k2023

Cite this paper

@article{xiong2023doctorglm,
  title  = {DoctorGLM: Fine-tuning your Chinese Doctor is not a Herculean Task},
  author = {Honglin Xiong and Sheng Wang and Yitao Zhu and Zihao Zhao and Yuxiao Liu and Linlin Huang and Qian Wang and Dinggang Shen},
  year   = {2023},
  eprint = {2304.01097},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.01097v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.01097.svg)](https://codeofpaper.com/paper/2304.01097)

Discussion