Coverage Embedding Models for Neural Machine Translation

In this paper, we enhance the attention-based neural machine translation by adding an explicit coverage embedding model to alleviate issues of repeating and dropping translations in NMT. For each source word, our model starts with a full coverage embedding vector, and then keeps updating it with a gated recurrent unit as the translation goes. All the initialized coverage embeddings and updating matrix are learned in the training procedure. Experiments on the large-scale Chineseto-English task show that our enhanced model improves the translation quality significantly on various test sets over the strong large vocabulary NMT system.