Facebook AI's WAT19 Myanmar-English Translation Task Submission
This work addresses translation quality for low-resource languages like Myanmar, though it is incremental as it applies existing techniques to a specific dataset.
The paper tackled the Myanmar-English translation task by exploring methods like self-training and back-translation to leverage monolingual data, achieving a first-place ranking with over 8 BLEU points above the second best system.
This paper describes Facebook AI's submission to the WAT 2019 Myanmar-English translation task. Our baseline systems are BPE-based transformer models. We explore methods to leverage monolingual data to improve generalization, including self-training, back-translation and their combination. We further improve results by using noisy channel re-ranking and ensembling. We demonstrate that these techniques can significantly improve not only a system trained with additional monolingual data, but even the baseline system trained exclusively on the provided small parallel dataset. Our system ranks first in both directions according to human evaluation and BLEU, with a gain of over 8 BLEU points above the second best system.