mteb/

NorwegianCourtsBitextMining

NorwegianCourtsBitextMining An MTEB dataset Massive Text Embedding Benchmark

Nynorsk and Bokmål parallel corpus from Norwegian courts. Norwegian courts have two standardised written languages. Bokmål is a variant closer to Danish, while Nynorsk was created to resemble regional dialects of Norwegian.

Task category t2t

Domains Legal, Written

Reference https://opus.nlpl.eu/index.php

You can evaluate an embedding model on this… See the full description on the dataset page: https://huggingface.co/datasets/mteb/NorwegianCourtsBitextMining.