2 datasets found

The dataset used in this paper is a multilingual parallel corpus, specifically the Watchtower corpus (WTC), which is a collection of multilingual sentences.

The WikiMatrix dataset is a multilingual dataset that contains parallel texts between English and other languages.