Corpus with Low-resrouce for Neural Machine Translation

These datasets were used in article "Improving Low-Resource Neural Machine Translation by Semantic Distance Augmentation".They are traing dataset, dev-set and test-set for each translation tasks: IWSLT14(De-En), IWSLT15(En-Vi), WMT18 (En-Tr), and NC11 (En-De).The original url to download these dataset can be found in the following address: (https://www.statmt.org/wmt18/, https://workshop2015.iwslt.org/, https://opus.nlpl.eu/legacy/News-Commentary-v11.php and https://huggingface.co/datasets/bbaaaa/iwslt14-de-en/blob/main/data/de-en.zip)