A neural interlingua for multilingual machine translation.

We incorporate an explicit neural interlingua into a multilingual encoder-decoder neural machine translation (NMT) architecture. We demonstrate that our model learns a true interlingua by performing direct zero-shot translation (without using pivot translation), and by using the interlingual sentence embeddings to train an English Yelp review classifier that, through the mediation of the interlingua, can also classify French and German reviews. Furthermore, we show that, despite using a smaller number of parameters than a pairwise collection of bilingual NMT models, our interlingual approach produces comparable BLEU scores for each language pair in WMT15.