A Neural Network based Approach to Automatic Post-Editing

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 first-stage MT system. Our APE system (AP E Sym) 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 (AP E Rerank) of the n-best translations from the phrase-based APE and AP E Sym systems provides further substantial improvements over the symmetric neural APE model. Human evaluation confirms that the AP E Rerank generated PE translations improve on the previous best neural APE system at WMT 2016.

Neural networks at the human mind: translator's hidden layers

“…Our aim is to train students either to translate or post-edit specialized texts 1 . "MT systems primarily make two types of errors -lexical and reordering errors" [6] and our students might be well prepared to go through the lexical level with a hidden conceptual background.…”

Multi-source Neural Automatic Post-Editing: FBK’s participation in the WMT 2017 APE shared task

“…Jointly learning from both source and translation has been previously proved to be effective in (Béchara et al, 2011;Chatterjee et al, 2015b). Such works, however, exploit the idea of a "joint representation" of the input mainly in the statistical phrase-based APE framework while, within the neural paradigm, recent prior work mostly focuses on single-source systems (Pal et al, 2016a;Junczys-Dowmunt and Grundkiewicz, 2016;Pal et al, 2017). The only exception, to the best of our knowledge, is the approach of Libovický et al (2016), who developed a multi-source neural APE system.…”

Neural Automatic Post-Editing Using Prior Alignment and Reranking

“…Usually APE tasks focus on systematic errors made by first stage MT systems, acting as an effective remedy to some of the inaccuracies in raw MT output. APE approaches cover a wide methodological range such as SMT techniques (Simard et al, 2007a;Simard et al, 2007b;Chatterjee et al, 2015;Pal et al, 2015;Pal et al, 2016d) real time integration of post-editing in MT (Denkowski, 2015), rule-based approaches to APE (Mareček et al, 2011;Rosa et al, 2012), neural APE (JunczysDowmunt and Grundkiewicz, 2016;Pal et al, 2016b), multi-engine and multi-alignment APE (Pal et al, 2016a), etc.…”