GMAI-MMBench
收藏arXiv2025-09-30 收录
下载链接:
https://huggingface.co/datasets/opengvlab/gmai-mmbench
下载链接
链接失效反馈官方服务:
资源简介:
该数据集为评估通用医学人工智能图像任务提供了一个基准。在视觉问答(VQA)任务中,Infi-Med模型在此基准上相较于基线模型显示出可衡量的改进。
This dataset provides a benchmark for evaluating general medical artificial intelligence image-related tasks. For the visual question answering (VQA) task, the Infi-Med model demonstrates measurable improvements over baseline models on this benchmark.



