ai4bharat/IndicMSMARCO

IndicMSMARCO是一个面向印度语言的多元语言信息检索基准数据集，包含精心挑选的查询和对应的高质量翻译段落。支持13种印度语言，适用于信息检索/段落排名任务，具有多样化的主题和查询复杂度。IndicMSMARCO is a comprehensive multilingual variant of MS MARCO tailored for Indian languages, featuring carefully selected queries and corresponding passages with high-quality translations. It supports 13 Indian languages and is suitable for information retrieval/passage ranking tasks with diverse topics and query complexities.