DivEMT: Neural Machine Translation Post-Editing Effort Across Typologically Diverse Languages

Gabriele Sarti*, Arianna Bisazza, Ana Guerberof, Antonio Toral Ruiz

*Bijbehorende auteur voor dit werk

OnderzoeksoutputAcademicpeer review

30 Downloads (Pure)


We introduce DivEMT, the first publicly available post-editing study of Neural Machine Translation (NMT) over a typologically diverse set of target languages. Using a strictly controlled setup, 18 professional translators were instructed to translate or post-edit the same set of English documents into Arabic, Dutch, Italian, Turkish, Ukrainian, and Vietnamese. During the process, their edits, keystrokes, editing times and pauses were recorded, enabling an in-depth, cross-lingual evaluation of NMT quality and post-editing effectiveness. Using this new dataset, we assess the impact of two state-of-the-art NMT systems, Google Translate and the multilingual mBART-50 model, on translation productivity. We find that post-editing is consistently faster than translation from scratch. However, the magnitude of productivity gains varies widely across systems and languages, highlighting major disparities in post-editing effectiveness for languages at different degrees of typological relatedness to English, even when controlling for system architecture and training data size. We publicly release the complete dataset including all collected behavioral data, to foster new research on the translation capabilities of NMT systems for typologically diverse languages.
Originele taal-2English
TitelProceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
RedacteurenYoav Goldberg, Zornitsa Kozareva, Yue Zhang
Plaats van productieAbu Dhabi, United Arab Emirates
UitgeverijAssociation for Computational Linguistics (ACL)
Aantal pagina's22
StatusPublished - dec.-2022
EvenementThe 2022 Conference on Empirical Methods in Natural Language Processing - Abu Dhabi National Exhibition Centre, Abu Dhabi, United Arab Emirates
Duur: 7-dec.-202211-dec.-2022


ConferenceThe 2022 Conference on Empirical Methods in Natural Language Processing
Verkorte titelEMNLP '22
Land/RegioUnited Arab Emirates
StadAbu Dhabi
Internet adres

Citeer dit