Neural automatic post-editing using prior alignment and reranking

Abstract

We present a second-stage machine translation (MT) system based on a neural machine translation (NMT) approach to automatic post-editing (APE) that improves the translation quality provided by a firststage MT system. Our APE system (APESym) is an extended version of an attention based NMT model with bilingual symmetry employing bidirectional models, mt ! pe and pe ! mt. APE translations produced by our system show statistically significant improvements over the first-stage MT, phrase-based APE and the best reported score on the WMT 2016 APE dataset by a previous neural APE system. Re-ranking (APERerank) of the n-best translations from the phrase-based APE and APESym systems provides further substantial improvements over the symmetric neural APE model. Human evaluation confirms that the APERerank generated PE translations improve on the previous best neural APE system at WMT 2016.