antonioloison/clean-colpali-dataset

该数据集包括两个部分：corpus和queries。corpus部分包含图像和与之相关的原始文本以及语言信息，适用于图像和文本相关任务。queries部分包含查询文本，正例 passages 索引，负例 passages 索引和原始文本信息，适用于文本检索或问答系统相关任务。数据集分为训练集和测试集（对于queries部分），提供了充足的示例以供模型训练和评估。

The dataset consists of two parts: corpus and queries. The corpus part includes images, associated original text, and language information, suitable for image and text-related tasks. The queries part includes query text, indices of positive passages, indices of negative passages, and original text information, suitable for text retrieval or question-answering system tasks. The dataset is split into training and test sets (for the queries part), providing a substantial number of examples for model training and evaluation.