Unsupervised Machine Translation Using Monolingual Corpora Only
This work addresses the challenge of translation for low-resource language pairs by enabling unsupervised learning, which is a significant advancement over methods requiring parallel data.
The paper tackles the problem of machine translation without any parallel data by proposing a model that maps monolingual sentences from two languages into a shared latent space and learns to reconstruct them, achieving BLEU scores of 32.8 and 15.1 on Multi30k and WMT English-French datasets.
Machine translation has recently achieved impressive performance thanks to recent advances in deep learning and the availability of large-scale parallel corpora. There have been numerous attempts to extend these successes to low-resource language pairs, yet requiring tens of thousands of parallel sentences. In this work, we take this research direction to the extreme and investigate whether it is possible to learn to translate even without any parallel data. We propose a model that takes sentences from monolingual corpora in two different languages and maps them into the same latent space. By learning to reconstruct in both languages from this shared feature space, the model effectively learns to translate without using any labeled data. We demonstrate our model on two widely used datasets and two language pairs, reporting BLEU scores of 32.8 and 15.1 on the Multi30k and WMT English-French datasets, without using even a single parallel sentence at training time.