antonioloison/mmarco-corpus-queries

该数据集包含了德语、英语、西班牙语、法语和意大利语五种语言的语料库和查询数据。语料库中的每条记录包含一个唯一ID、文本内容和语言类型。查询数据包括查询文本、正例段落ID序列、反例段落ID序列、语言类型、翻译后的查询文本和翻译语言列表。数据集分为训练集和测试集。

The dataset includes corpora and query data in five languages: German (corpus-de), English (corpus-en), Spanish (corpus-es), French (corpus-fr), and Italian (corpus-it). Each corpus entry contains a unique ID, text content, and language type. The query data includes the query text, sequence of positive passage IDs, sequence of negative passage IDs, language type, translated query text, and translation language list. The dataset is split into training and test sets.