antonioloison/filtered-llamaindex-with-translations

该数据集包含两部分：corpus和queries。corpus部分包含图片及其对应的语言文字描述；queries部分包含查询相关的信息，如正例文本段落、答案可能性、原始文本、原始语言、原始查询、翻译语言和翻译后的查询。数据集分为训练集和测试集，可用于训练和评估相关模型。

The dataset consists of two parts: corpus and queries. The corpus part includes images and their corresponding language descriptions; the queries part includes information related to queries, such as positive passages, answerabilities, original text, original language, original query, translation languages, and translated queries. The dataset is split into training and test sets for training and evaluating relevant models.