mlx-community/medfit-dataset

MEDFIT医疗问答数据集包含6444个独特的医疗相关问答对，旨在为医疗聊天应用程序微调语言模型。该数据集专门为MEDFIT-LLM研究项目设计，该项目专注于小规模语言模型在医疗应用领域的领域专注微调。数据集经过合成数据生成、领域特定策划、去重和医疗准确性质量控制。它适用于聊天机器人训练和评估，并已在同行评审的研究中显示显著的改进。

The MEDFIT Medical QA Dataset contains 6,444 unique healthcare-related question-answer pairs designed for fine-tuning language models for medical chatbot applications. Curated for the MEDFIT-LLM research project, the dataset focuses on domain-specific content, undergoes synthetic data generation, domain-specific curation, deduplication, and medical accuracy quality control. It is formatted for chatbot training and evaluation and has demonstrated significant improvements in peer-reviewed research.