A Transformer-Based Multi-Source Automatic Post-Editing System

This paper presents our English–German Automatic Post-Editing (APE) system submitted to the APE Task organized at WMT 2018 (Chatterjee et al., 2018). The proposed model is an extension of the transformer architecture: two separate self-attention-based encoders encode the machine translation output (mt) and the source (src), followed by a joint encoder that attends over a combination of these two encoded sequences (encsrc and encmt) for generating the post-edited sentence. We compare this multi-source architecture (i.e, {src, mt} → pe) to a monolingual transformer (i.e., mt → pe) model and an ensemble combining the multi-source {src, mt} → pe and singlesource mt → pe models. For both the PBSMT and the NMT task, the ensemble yields the best results, followed by the multi-source model and last the singlesource approach. Our best model, the ensemble, achieves a BLEU score of 66.16 and 74.22 for the PBSMT and NMT task, respectively.

[1]  Marcin Junczys-Dowmunt,et al.  Log-linear Combinations of Monolingual and Bilingual Neural Machine Translation Models for Automatic Post-Editing , 2016, WMT.

[2]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Philipp Koehn,et al.  Moses: Open Source Toolkit for Statistical Machine Translation , 2007, ACL.

[4]  Santanu Pal,et al.  Multi-source Neural Automatic Post-Editing: FBK’s participation in the WMT 2017 APE shared task , 2017, WMT.

[5]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[6]  Josef van Genabith,et al.  Multi-Engine and Multi-Alignment Based Automatic Post-Editing and its Impact on Translation Productivity , 2016, COLING.

[7]  Kevin Knight,et al.  Automated Postediting of Documents , 1994, AAAI.

[8]  Yoshua Bengio,et al.  Neural Machine Translation by Jointly Learning to Align and Translate , 2014, ICLR.

[9]  Josef van Genabith,et al.  Neural Automatic Post-Editing Using Prior Alignment and Reranking , 2017, EACL.

[10]  Philipp Koehn,et al.  Findings of the 2017 Conference on Machine Translation (WMT17) , 2017, WMT.

[11]  Yann Dauphin,et al.  Convolutional Sequence to Sequence Learning , 2017, ICML.

[12]  Marco Turchi,et al.  ESCAPE: a Large-scale Synthetic Corpus for Automatic Post-Editing , 2018, LREC.

[13]  Alon Lavie,et al.  Better Hypothesis Testing for Statistical Machine Translation: Controlling for Optimizer Instability , 2011, ACL.

[14]  Matteo Negri,et al.  Findings of the WMT 2018 Shared Task on Automatic Post-Editing , 2018, WMT.

[15]  Manuel Arcedillo,et al.  Living on the edge: productivity gain thresholds in machine translation evaluation metrics , 2015, MTSUMMIT.

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Ondrej Dusek,et al.  DEPFIX: A System for Automatic Correction of Czech MT Outputs , 2012, WMT@NAACL-HLT.

[18]  Marion Weller,et al.  Exploring the Planet of the APEs: a Comparative Study of State-of-the-art Methods for MT Automatic Post-Editing , 2015, ACL.

[19]  Marco Turchi,et al.  Online Automatic Post-editing for MT in a Multi-Domain Translation Environment , 2017, EACL.

[20]  Francisco Casacuberta,et al.  Statistical Post-Editing of a Rule-Based Machine Translation System , 2009, NAACL.

[21]  Josef van Genabith,et al.  Statistical Post-Editing for a Statistical MT System , 2011, MTSUMMIT.

[22]  Jindřich Helcl,et al.  CUNI System for WMT16 Automatic Post-Editing and Multimodal Translation Tasks , 2016, WMT.

[23]  Josef van Genabith,et al.  USAAR: An Operation Sequential Model for Automatic Statistical Post-Editing , 2016, WMT.

[24]  Carla Parra Escartín,et al.  Machine translation evaluation made fuzzier: a study on post-editing productivity and evaluation metrics in commercial settings , 2015, MTSUMMIT.

[25]  Ondrej Bojar,et al.  CUNI System for WMT17 Automatic Post-Editing Task , 2017, WMT.

[26]  Roland Kuhn,et al.  Rule-Based Translation with Statistical Phrase-Based Post-Editing , 2007, WMT@ACL.

[27]  Karin M. Verspoor,et al.  Findings of the 2016 Conference on Machine Translation , 2016, WMT.

[28]  Josef van Genabith,et al.  UdS-Sant: English–German Hybrid Machine Translation System , 2015, WMT@EMNLP.

[29]  Rico Sennrich,et al.  Neural Machine Translation of Rare Words with Subword Units , 2015, ACL.

[30]  Michel Simard,et al.  Statistical Phrase-Based Post-Editing , 2007, NAACL.

[31]  Jan Niehues,et al.  Pre-Translation for Neural Machine Translation , 2016, COLING.

[32]  Josef van Genabith,et al.  A Neural Network based Approach to Automatic Post-Editing , 2016, ACL.

[33]  Johann Roturier,et al.  Deploying Novel MT Technology to Raise the Bar for Quality at Symantec: Key Advantages and Challenge , 2009, MTSUMMIT.

[34]  Marcin Junczys-Dowmunt,et al.  The AMU-UEdin Submission to the WMT 2017 Shared Task on Automatic Post-Editing , 2017, WMT.