Reference Language based Unsupervised Neural Machine Translation
This work addresses performance limitations in unsupervised machine translation for scenarios lacking direct source-target parallel data, though it is incremental as it builds on existing pivot translation ideas.
The authors tackled the problem of unsatisfactory performance in unsupervised neural machine translation (UNMT) due to vague training clues, by proposing a reference language-based framework (RUNMT) that uses a parallel corpus shared only with the source language to enhance reconstruction training. Experimental results show improved translation quality over a strong baseline using only one auxiliary language.
Exploiting a common language as an auxiliary for better translation has a long tradition in machine translation and lets supervised learning-based machine translation enjoy the enhancement delivered by the well-used pivot language in the absence of a source language to target language parallel corpus. The rise of unsupervised neural machine translation (UNMT) almost completely relieves the parallel corpus curse, though UNMT is still subject to unsatisfactory performance due to the vagueness of the clues available for its core back-translation training. Further enriching the idea of pivot translation by extending the use of parallel corpora beyond the source-target paradigm, we propose a new reference language-based framework for UNMT, RUNMT, in which the reference language only shares a parallel corpus with the source, but this corpus still indicates a signal clear enough to help the reconstruction training of UNMT through a proposed reference agreement mechanism. Experimental results show that our methods improve the quality of UNMT over that of a strong baseline that uses only one auxiliary language, demonstrating the usefulness of the proposed reference language-based UNMT and establishing a good start for the community.