mteb/NeuCLIR2022RetrievalHardNegatives

NeuCLIR2022RetrievalHardNegatives是一个多语言文本检索数据集，包含Farsi、Russian和Chinese三种语言的文本数据，用于训练和评估文本检索模型。数据集由mteb/neuclir-2022和mteb/neuclir-2022-hard-negatives两个来源的数据组成，包含了测试集的语料库、查询和相关性评分信息。NeuCLIR2022RetrievalHardNegatives is a multilingual text retrieval dataset containing texts in Farsi, Russian, and Chinese, designed for training and evaluating text retrieval models. The dataset is composed of data from mteb/neuclir-2022 and mteb/neuclir-2022-hard-negatives, including the test sets corpus, queries, and relevance scores.