Youku-mPLUG

为更好地推动多模态技术在华人社区的应用和发展，达摩院与优酷联合发布了业界首个开源的大规模中文视频文本数据集优酷-mPLUG: 包括从优酷海量视频库中挖掘出的数百万条高质量视频作为多模态预训练数据集，以及多个下游多模态基准 (视频检索、视频描述、类别预测)。此外，我们为基准测试提供了不同的多模态预训练模型，以加速多模态预训练技术的应用和发展。

To better promote the application and development of multimodal technologies in the Chinese-speaking community, Damo Academy and Youku jointly released Youku-mPLUG, the industry's first open-source large-scale Chinese video-text dataset. This dataset includes millions of high-quality videos mined from Youku's massive video repository for multimodal pre-training, as well as multiple downstream multimodal benchmarks covering video retrieval, video captioning, and category prediction. Additionally, we provide various multimodal pre-trained models for benchmark testing to accelerate the application and advancement of multimodal pre-training technologies.

搜集汇总

背景与挑战

背景概述

Youku-mPLUG是由阿里巴巴达摩院与优酷联合发布的大规模中文视频文本数据集，包含数百万条高质量视频用于多模态预训练。该数据集支持多个下游基准任务，如视频检索、视频描述和类别预测，并提供了预训练模型以加速多模态技术的应用和发展。