TL;DR
This paper details LIUM's neural machine translation systems for multiple language pairs in WMT17, utilizing BPE, ensembling, and back-translation to improve translation quality, with notable success in English-Turkish.
Contribution
Introduction of BPE-based neural translation systems with ensembling and back-translation techniques for multiple language pairs, achieving competitive results.
Findings
Back-translation improves translation quality.
Ensembling enhances system performance.
Post-deadline English-Turkish system surpasses previous best by +1.6 BLEU.
Abstract
This paper describes LIUM submissions to WMT17 News Translation Task for English-German, English-Turkish, English-Czech and English-Latvian language pairs. We train BPE-based attentive Neural Machine Translation systems with and without factored outputs using the open source nmtpy framework. Competitive scores were obtained by ensembling various systems and exploiting the availability of target monolingual corpora for back-translation. The impact of back-translation quantity and quality is also analyzed for English-Turkish where our post-deadline submission surpassed the best entry by +1.6 BLEU.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
