Non-Autoregressive Neural Machine Translation with Enhanced Decoder Input
This work addresses the trade-off between speed and accuracy in machine translation for applications requiring fast inference, though it is incremental as it builds on existing NAT methods.
The paper tackles the accuracy gap in non-autoregressive neural machine translation by enhancing decoder inputs, achieving improvements of 5.11 BLEU on WMT14 English-German and 4.72 BLEU on WMT16 English-Romanian tasks.
Non-autoregressive translation (NAT) models, which remove the dependence on previous target tokens from the inputs of the decoder, achieve significantly inference speedup but at the cost of inferior accuracy compared to autoregressive translation (AT) models. Previous work shows that the quality of the inputs of the decoder is important and largely impacts the model accuracy. In this paper, we propose two methods to enhance the decoder inputs so as to improve NAT models. The first one directly leverages a phrase table generated by conventional SMT approaches to translate source tokens to target tokens, which are then fed into the decoder as inputs. The second one transforms source-side word embeddings to target-side word embeddings through sentence-level alignment and word-level adversary learning, and then feeds the transformed word embeddings into the decoder as inputs. Experimental results show our method largely outperforms the NAT baseline~\citep{gu2017non} by $5.11$ BLEU scores on WMT14 English-German task and $4.72$ BLEU scores on WMT16 English-Romanian task.