Jinapeng/MMLM_EVAL
收藏Hugging Face2025-03-29 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/Jinapeng/MMLM_EVAL
下载链接
链接失效反馈官方服务：
资源简介：
Massive Multimodal Embedding Benchmark是一个用于评估多模态嵌入模型性能的数据集集合，包含4个元任务和36个子数据集，每个数据集有1000个示例，用于训练和评估模型对图像和文本组合的处理能力。
Massive Multimodal Embedding Benchmark is a collection of datasets designed for evaluating the performance of multimodal embedding models, comprising 4 meta-tasks and 36 sub-datasets, each with 1000 examples for training and evaluation of the models ability to process combinations of images and text.
提供机构：
Jinapeng
