openbmb/VisRAG-Ret-Test-SlideVQA

这是一个基于SlideVQA数据集的幻灯片演示文稿的VQA数据集。它包含三个配置：语料库，qrels和查询。语料库配置包含图像和文本数据（语料库-id），qrels配置包括查询与语料项之间的相关性评分，查询配置则包含实际的查询以及可能的答案和选项。数据集为训练集提供了示例数量和字节数。数据集的结构包括每种配置的特定数据文件。

This is a VQA dataset based on Slide Decks from the SlideVQA dataset. It includes three configurations: corpus, qrels, and queries. The corpus configuration contains image and text data (corpus-id), the qrels configuration includes relevance scores between queries and corpus items, and the queries configuration holds the actual queries with possible answers and options. The dataset provides the number of examples and bytes for the train split. The dataset is structured with specific data files for each configuration.