Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Dataset

TL;DR

This paper introduces CMExam, a large-scale Chinese medical exam dataset with annotations, to evaluate and analyze large language models' performance in medical question answering, revealing significant gaps compared to human experts.

Contribution

The paper presents the first comprehensive Chinese medical exam dataset with annotations and benchmarks LLMs, providing insights into their capabilities and limitations in medical QA.

Findings

GPT-4 achieved 61.6% accuracy on CMExam.

LLMs lag behind human accuracy of 71.6%.

Finetuning improves LLM reasoning but still falls short.

Abstract

Recent advancements in large language models (LLMs) have transformed the field of question answering (QA). However, evaluating LLMs in the medical field is challenging due to the lack of standardized and comprehensive datasets. To address this gap, we introduce CMExam, sourced from the Chinese National Medical Licensing Examination. CMExam consists of 60K+ multiple-choice questions for standardized and objective evaluations, as well as solution explanations for model reasoning evaluation in an open-ended manner. For in-depth analyses of LLMs, we invited medical professionals to label five additional question-wise annotations, including disease groups, clinical departments, medical disciplines, areas of competency, and question difficulty levels. Alongside the dataset, we further conducted thorough experiments with representative LLMs and QA algorithms on CMExam. The results show that GPT-4 had the best accuracy of 61.6% and a weighted F1 score of 0.617. These results highlight a great disparity when compared to human accuracy, which stood at 71.6%. For explanation tasks, while LLMs could generate relevant reasoning and demonstrate improved performance after finetuning, they fall short of a desired standard, indicating ample room for improvement. To the best of our knowledge, CMExam is the first Chinese medical exam dataset to provide comprehensive medical annotations. The experiments and findings of LLM evaluation also provide valuable insights into the challenges and potential solutions in developing Chinese medical QA systems and LLM evaluation pipelines. The dataset and relevant code are available at https://github.com/williamliujl/CMExam.