universitytehran/PerMed-MM
收藏Hugging Face2025-11-09 更新2025-11-15 收录
下载链接：
https://hf-mirror.com/datasets/universitytehran/PerMed-MM
下载链接
链接失效反馈官方服务：
资源简介：
PerMed-MM是一个针对波斯语医学问题回答的视觉语言模型（VLMs）评估的多模态、多专业基准数据集。该数据集包含来自伊朗国家医学委员会考试的733个多项选择题，每个问题配有一到五个临床相关图像，总计944个图像。这个基准跨越了46个医学专业，并涵盖了一系列的视觉模态，包括放射影像、组织病理学切片、皮肤科照片和心电图波形。
PerMed-MM is the first multimodal, multi-specialty benchmark for evaluating Vision Language Models (VLMs) on Persian medical question answering. The dataset comprises 733 multiple-choice questions sourced from the Iranian National Medical Board Exams. Each question is paired with one to five clinically relevant images, totaling 944 images. The benchmark spans 46 medical specialties and covers a wide range of visual modalities, including radiographic images, histopathology slides, dermatologic photographs, and ECG waveforms.
提供机构：
universitytehran
