Abstract
Document-level machine translation focuses on the translation of entire documents from a source to a target language. It is widely regarded as a challenging task since the translation of the individual sentences in the document needs to retain aspects of the discourse at document level. However, document-level translation models are usually not trained to explicitly ensure discourse quality. Therefore, in this paper we propose a training approach that explicitly optimizes two established discourse metrics, lexical cohesion (LC) and coherence (COH), by using a reinforcement learning objective. Experiments over four different language pairs and three translation domains have shown that our training approach has been able to achieve more cohesive and coherent document translations than other competitive approaches, yet without compromising the faithfulness to the reference translation. In the case of the Zh-En language pair, our method has achieved an improvement of 2.46 percentage points (pp) in LC and 1.17 pp in COH over the runner-up, while at the same time improving 0.63 pp in BLEU score and 0.47
Original language | English |
---|---|
Title of host publication | COLING 2020 |
Subtitle of host publication | The 28th International Conference on Computational Linguistics, Proceedings of the Conference |
Editors | Nuria Bel, Chengquing Zong |
Place of Publication | Stroudsburg PA USA |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 4467–4482 |
Number of pages | 16 |
ISBN (Electronic) | 9781952148279 |
DOIs | |
Publication status | Published - 2020 |
Event | International Conference on Computational Linguistics 2020 - Virtual, Barcelona, Spain Duration: 8 Dec 2020 → 13 Dec 2020 Conference number: 28th https://coling2020.org (Website) https://www.aclweb.org/anthology/volumes/2020.coling-main/ (Proceedings) |
Conference
Conference | International Conference on Computational Linguistics 2020 |
---|---|
Abbreviated title | COLING 2020 |
Country/Territory | Spain |
City | Barcelona |
Period | 8/12/20 → 13/12/20 |
Internet address |
|