CoIR-Retrieval/apps

该数据集包含三个配置：语料库、默认和查询。每个配置具有特定特征，如_id、partition、text、language和meta_information。语料库和查询配置具有相似的特征，包括title字段。默认配置针对特定任务，具有query-id、corpus-id和score等特征。数据集分为不同的部分，例如，默认配置的train和test部分，以及语料库和查询配置的单个corpus部分。还提供了每个部分的大小和示例数量。另外，还介绍了如何使用MTEB评估框架和给定模型对各种任务进行评估。

The dataset consists of three configurations: corpus, default, and queries. Each configuration has specific features such as _id, partition, text, language, and meta_information. The corpus and queries configurations share similar features, including a title field. The default configuration is tailored for a specific task with features like query-id, corpus-id, and score. The dataset is split into different parts such as the train and test splits for the default configuration, and a single corpus split for the corpus and queries configurations. The file sizes and number of examples for each split are also provided. Additionally, the employment of the MTEB evaluation framework with a given model for assessment on various tasks is described.