eherra/

EuroPIRQ-retrieval

	EuroPIRQ: European Parallel Information Retrieval Queries




	
	Dataset Details

The EuroPIRQ retrieval dataset is a multilingual collection designed for evaluating retrieval and cross-lingual retrieval tasks. Dataset contains 10,000 parallel passages & 100 parallel queries (synthetic) in three languages: English🇬🇧, Portuguese🇵🇹, and Finnish🇫🇮, constructed from the European Union's DGT-Acquis corpus.

Languages: English (en), Portuguese (pt), Finnish (fi) Domain: Legal /… See the full description on the dataset page: https://huggingface.co/datasets/eherra/EuroPIRQ-retrieval.