Key points are not available for this paper at this time.
This paper proposes three distortion models to explicitly incorporate the word reordering knowledge into attention-based Neural Machine Translation (NMT) for further improving translation performance.
Zhang et al. (Sun,) studied this question.