Data Diversification: A Simple Strategy For Neural Machine Translation

TL;DR

The paper presents Data Diversification, a straightforward data augmentation technique for neural machine translation that improves performance by merging original data with predictions from multiple models, without extra data or computational costs.

Contribution

It introduces a novel data diversification strategy that enhances NMT performance by leveraging model predictions, outperforming existing methods like knowledge distillation and dual learning.

Findings

Achieves state-of-the-art BLEU scores on WMT'14 English-German and English-French tasks.

Significantly improves translation quality on multiple low-resource and IWSLT tasks.

More effective than knowledge distillation and dual learning methods.

Abstract

We introduce Data Diversification: a simple but effective strategy to boost neural machine translation (NMT) performance. It diversifies the training data by using the predictions of multiple forward and backward models and then merging them with the original dataset on which the final NMT model is trained. Our method is applicable to all NMT models. It does not require extra monolingual data like back-translation, nor does it add more computations and parameters like ensembles of models. Our method achieves state-of-the-art BLEU scores of 30.7 and 43.7 in the WMT'14 English-German and English-French translation tasks, respectively. It also substantially improves on 8 other translation tasks: 4 IWSLT tasks (English-German and English-French) and 4 low-resource translation tasks (English-Nepali and English-Sinhala). We demonstrate that our method is more effective than knowledge distillation and dual learning, it exhibits strong correlation with ensembles of models, and it trades perplexity off for better BLEU score. We have released our source code at https://github.com/nxphi47/data_diversification