MMedBench 多语言医学能力测试基准数据集

MMedBench 是一个全面多语言医学能力测试基准数据集，由上海交通大学人工智能学院智慧医疗团队于 2024 年开发，论文成果为「Towards building multilingual language model for medicine」。它旨在评估医学领域多语言模型的发展，涵盖了 6 种语言和 21 种医学子领域。 MMedBench 的所有问题直接来源于各国的医学考试题库，确保了评测的准确性和可靠性，避免了由于不同国家医疗实践指南差异导致的诊断理解偏差。MMedBench is a comprehensive multilingual medical proficiency test benchmark dataset, developed by the Smart Healthcare Team of the School of Artificial Intelligence, Shanghai Jiao Tong University in 2024, with its corresponding paper titled "Towards Building a Multilingual Language Model for Medicine". It aims to evaluate the development of multilingual medical models, covering 6 languages and 21 medical subfields. All questions in MMedBench are directly sourced from medical examination question banks across various countries, ensuring the accuracy and reliability of the evaluation while avoiding diagnostic understanding biases caused by disparities in medical practice guidelines among different nations.

搜集汇总

背景与挑战

背景概述

MMedBench是一个由上海交通大学于2024年开发的多语言医学能力测试基准数据集，涵盖6种语言和21个医学子领域，问题源自各国医学考试题库以确保准确性。它通过选择准确率和解释合理性两个维度评估模型，测试显示其模型性能超越同级别开源模型并与GPT-4相当，且已开源以推动全球医疗AI研究。