TIGER-Lab/MMEB_Raw_Video

MMEB-V2是在原始MMEB数据集基础上的扩展，它包含了五个任务：视频检索、瞬间检索、视频分类、视频问答以及视觉文档检索。这个全面的任务集使得多模态嵌入模型能够在静态、时态和结构化视觉数据环境中进行健壮的评估。Hugging Face仓库中包含MMEB-V2使用的原始视频文件，以供需要时使用。

MMEB-V2 builds upon the original MMEB dataset, including five tasks: Video Retrieval, Moment Retrieval, Video Classification, Video Question Answering, and Visual Document Retrieval. This comprehensive suite allows for robust evaluation of multimodal embedding models across static, temporal, and structured visual data settings. The Hugging Face repository contains the raw video files used in MMEB-V2, provided for cases where they are necessary.