NTT-hil-insight/OpenDocVQA

OpenDocVQA 是一个开放的文档视觉问答数据集集合，包含了多种文档类型和格式。该数据集旨在训练和评估 VDocRAG 框架。任务分为两个阶段：视觉文档检索和文档视觉问答。数据集支持英语，包含多个子数据集，每个子数据集都有特定的许可条件。README 文件还提供了数据实例示例、数据字段描述、数据集统计数据以及关于许可和引用的附加信息。

OpenDocVQA is a collection of open-domain document visual question answering datasets, encompassing diverse document types and formats. The dataset is designed for training and evaluating the VDocRAG framework. The task is divided into two stages: visual document retrieval and document visual question answering. The dataset supports English language and includes multiple sub-datasets with specific licensing conditions. The README also includes a data instance example, data fields description, dataset statistics, and additional information about licenses and citations.