allenai/Molmo2-CapEval
收藏Hugging Face2025-12-16 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/allenai/Molmo2-CapEval
下载链接
链接失效反馈官方服务：
资源简介：
Molmo2-CapEval是一个包含多个标注者为每个视频提供的非常长且详细字幕的数据集。它可用于测试视觉语言模型生成字幕的能力。数据集是Molmo2数据集集合的一部分，用于测试Molmo2系列模型。视频来源于Vimeo、Ego4D和BDD100K，以视频ID的形式存储，需要单独下载。数据集的特征包括视频ID、来源、开始和结束时间、持续时间、原子语句、语句类别和聚合字幕。
Molmo2-CapEval is a dataset of very long, detailed video captions from multiple annotators per video. It can be used to test the caption capability of vision-language models. Molmo2-Cap is part of the Molmo2 dataset collection and was used to test the Molmo2 family of models. Videos are from Vimeo, Ego4D, and BDD100K; all stored as video ID that will need to be downloaded separately. The datasets features include video ID, source, start and end time, duration, atomic statements, statement categories, and aggregated caption.
提供机构：
allenai
