NIH-CARD/CARDBiomedBench
收藏Hugging Face2025-07-21 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/NIH-CARD/CARDBiomedBench
下载链接
链接失效反馈官方服务：
资源简介：
CARDBiomedBench是一个生物医学问答基准数据集，旨在评估大型语言模型在复杂生物医学任务上的性能。该数据集由一系列经过精心挑选的问题和答案对组成，涵盖了多个生物医学领域和推理类型，挑战模型在生物医学领域的深入理解和推理能力。
CARDBiomedBench is a biomedical question-answering benchmark designed to evaluate the performance of Large Language Models (LLMs) on complex biomedical tasks. It consists of a curated set of question-answer pairs covering various biomedical domains and reasoning types, challenging models to demonstrate deep understanding and reasoning capabilities in the biomedical field.
提供机构：
NIH-CARD
