whooray/mrtidy

MrTidyRetrieval An MTEB dataset Massive Text Embedding Benchmark Mr. TyDi is a multi-lingual benchmark dataset built on TyDi, covering eleven typologically diverse languages. It is designed for monolingual retrieval, specifically to evaluate ranking with learned dense representations. Task category t2t Domains Encyclopaedic, Written Reference huggingface.co How to evaluate on this task