Multi-task Learning for Multilingual Neural Machine Translation
This work addresses improving translation quality for multiple languages, including low-resource ones, but is incremental as it builds on existing multi-task learning and denoising techniques.
The authors tackled the problem of leveraging monolingual data to improve multilingual neural machine translation by proposing a multi-task learning framework, achieving significantly better results than bilingual models for both high- and low-resource languages and demonstrating efficacy in zero-shot setups.
While monolingual data has been shown to be useful in improving bilingual neural machine translation (NMT), effectively and efficiently leveraging monolingual data for Multilingual NMT (MNMT) systems is a less explored area. In this work, we propose a multi-task learning (MTL) framework that jointly trains the model with the translation task on bitext data and two denoising tasks on the monolingual data. We conduct extensive empirical studies on MNMT systems with 10 language pairs from WMT datasets. We show that the proposed approach can effectively improve the translation quality for both high-resource and low-resource languages with large margin, achieving significantly better results than the individual bilingual models. We also demonstrate the efficacy of the proposed approach in the zero-shot setup for language pairs without bitext training data. Furthermore, we show the effectiveness of MTL over pre-training approaches for both NMT and cross-lingual transfer learning NLU tasks; the proposed approach outperforms massive scale models trained on single task.