Improving Zero-shot Multilingual Neural Machine Translation for Low-Resource Languages

TL;DR

This paper enhances zero-shot multilingual NMT for low-resource languages by introducing target language tags and replacing beam search with random sampling, significantly improving translation quality and diversity.

Contribution

It proposes a tagged-multilingual NMT model with target language tags and an improved self-learning algorithm using random sampling, addressing key issues in zero-shot translation.

Findings

Tagged-multilingual NMT outperforms baseline by 9.41 BLEU points.

Improved self-learning algorithm enhances zero-shot translation quality.

Model demonstrates significant gains on Romanian-Italian translation tasks.

Abstract

Although the multilingual Neural Machine Translation(NMT), which extends Google's multilingual NMT, has ability to perform zero-shot translation and the iterative self-learning algorithm can improve the quality of zero-shot translation, it confronts with two problems: the multilingual NMT model is prone to generate wrong target language when implementing zero-shot translation; the self-learning algorithm, which uses beam search to generate synthetic parallel data, demolishes the diversity of the generated source language and amplifies the impact of the same noise during the iterative learning process. In this paper, we propose the tagged-multilingual NMT model and improve the self-learning algorithm to handle these two problems. Firstly, we extend the Google's multilingual NMT model and add target tokens to the target languages, which associates the start tag with the target language to ensure that the source language can be translated to the required target language. Secondly, we improve the self-learning algorithm by replacing beam search with random sample to increases the diversity of the generated data and makes it properly cover the true data distribution. Experimental results on IWSLT show that the adjusted tagged-multilingual NMT separately obtains 9.41 and 7.85 BLEU scores over the multilingual NMT on 2010 and 2017 Romanian-Italian test sets. Similarly, it obtains 9.08 and 7.99 BLEU scores on Italian-Romanian zero-shot translation. Furthermore, the improved self-learning algorithm shows its superiorities over the conventional self-learning algorithm on zero-shot translations.