Proceedings of the 2nd Workshop on Neural Machine Translation and Generation

12 

(1)ACL 2018. Neural Machine Translation and Generation. Proceedings of the Second Workshop. July 20, 2018 Melbourne, Australia.

(2) c 2018 The Association for Computational Linguistics. Apple and the Apple logo are trademarks of Apple Inc., registered in the U.S. and other countries.. Order copies of this and other ACL proceedings from: Association for Computational Linguistics (ACL) 209 N. Eighth Street Stroudsburg, PA 18360 USA Tel: +1-570-476-8006 Fax: +1-570-476-0860 acl@aclweb.org ISBN 978-1-948087-40-7. ii.

(3) Introduction Machine Translation and Generation focusing on Neural Machine Translation (NMT) technology. This workshop aims to cultivate research on the leading edge in neural machine translation and other aspects of machine translation, generation, and multilinguality that utilize neural models. In this year’s workshop we are extremely pleased to be able to host four invited talks from leading lights in the field, namely: Jacob Devlin Rico Sennrich, Jason Weston, and Yulia Tsvetkov. In addition this year’s workshop will feature a session devoted to a new shared task on efficient machine translation. We received a total of 25 submissions, and accepted 16 for inclusion in the workshop, an acceptance rate of 64%. Due to the large number of invited talks, and to encourage discussion, only the two papers selected for best paper awards will be presented orally, and the remainder will be presented in a single poster session. We would like to thank all authors for their submissions, and the program committee members for their valuable efforts in reviewing the papers for the workshop. We would also like to thank Amazon, Apple and Google for their generous sponsorship.. iii.

(4)

(5) Organizers: Alexandra Birch, (Edinburgh) Andrew Finch, (Apple) Thang Luong, (Google) Graham Neubig, (CMU) Yusuke Oda, (Google) Program Committee: Roee Aharoni, (Bar Ilan University) Joost Bastings, (University of Amsterdam) Yonatan Belinkov, (Qatar Computing Research Institute) Marine Carpuat, (University of Maryland) Boxing Chen, (National Research Council, Canada) Eunah Cho, (Amazon) Michael Denkowski, (Amazon) Kevin Duh, (JHU) Cong Duy Vu Hoang, (University of Melbourne) Markus Freitag, (Google) Isao Goto, (NHK) Jiatao Gu, (The University of Hong Kong) Barry Haddow, (Edinburgh) Sebastien Jean, (Montreal) Yuta Kikuchi, (Preferred Networks) Philipp Koehn, (Johns Hopkins University) Yannis Konstas, (Heriot-Watt University) Shumpei Kubosawa, (NEC) Mirella Lapata, (University of Edinburgh) Shujie Liu, (Microsoft) Lemao Liu, (Tencent AI Lab) Haitao Mi, (Ant Financial US) Hideya Mino, (NHK) Makoto Morishita, (Nara Institute of Science and Technology) Preslav Nakov, (QCRI) Hieu Pham, (Google) Alexander Rush, (Harvard) Abigail See, (Stanford) Rico Sennrich, (Edinburgh) Raphael Shu, (The University of Tokyo) Akihiro Tamura, (Ehime University) Rui Wang, (NICT) Xiaolin Wang, (NICT) Taro Watanabe, (Google) Xingxing Zhang, (University of Edinburgh). v.

(6) Invited Speakers: Jacob Devlin, (Google) Rico Sennrich, (Edinburgh) Yulia Tsvetkov, (CMU) Jason Weston, (Facebook). vi.

(7) Table of Contents. Findings of the Second Workshop on Neural Machine Translation and Generation Alexandra Birch, Andrew Finch, Minh-Thang Luong, Graham Neubig and Yusuke Oda . . . . . . . . . 1 A Shared Attention Mechanism for Interpretation of Neural Automatic Post-Editing Systems Inigo Jauregi Unanue, Ehsan Zare Borzeshi and Massimo Piccardi . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Iterative Back-Translation for Neural Machine Translation Vu Cong Duy Hoang, Philipp Koehn, Gholamreza Haffari and Trevor Cohn . . . . . . . . . . . . . . . . . . 18 Inducing Grammars with and for Neural Machine Translation Yonatan Bisk and Ke Tran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Regularized Training Objective for Continued Training for Domain Adaptation in Neural Machine Translation Huda Khayrallah, Brian Thompson, Kevin Duh and Philipp Koehn . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Controllable Abstractive Summarization Angela Fan, David Grangier and Michael Auli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Enhancement of Encoder and Attention Using Target Monolingual Corpora in Neural Machine Translation Kenji Imamura, Atsushi Fujita and Eiichiro Sumita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Document-Level Adaptation for Neural Machine Translation Sachith Sri Ram Kothur, Rebecca Knowles and Philipp Koehn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 On the Impact of Various Types of Noise on Neural Machine Translation Huda Khayrallah and Philipp Koehn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Bi-Directional Neural Machine Translation with Synthetic Parallel Data Xing Niu, Michael Denkowski and Marine Carpuat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Multi-Source Neural Machine Translation with Missing Data Yuta Nishimura, Katsuhito Sudoh, Graham Neubig and Satoshi Nakamura . . . . . . . . . . . . . . . . . . . 92 Towards one-shot learning for rare-word translation with external experts Ngoc-Quan Pham, Jan Niehues and Alexander Waibel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 NICT Self-Training Approach to Neural Machine Translation at NMT-2018 Kenji Imamura and Eiichiro Sumita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Fast Neural Machine Translation Implementation Hieu Hoang, Tomasz Dwojak, Rihards Krislauks, Daniel Torregrosa and Kenneth Heafield . . . . 116 OpenNMT System Description for WNMT 2018: 800 words/sec on a single-core CPU Jean Senellart, Dakun Zhang, Bo WANG, Guillaume KLEIN, Jean-Pierre Ramatchandirin, Josep Crego and Alexander Rush . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Marian: Cost-effective High-Quality Neural Machine Translation in C++ Marcin Junczys-Dowmunt, Kenneth Heafield, Hieu Hoang, Roman Grundkiewicz and Anthony Aue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129. vii.

(8)

(9) Workshop Program July 20, 2018 09:00–09:10. Welcome and Opening Remarks Findings of the Second Workshop on Neural Machine Translation and Generation Alexandra Birch, Andrew Finch, Minh-Thang Luong, Graham Neubig and Yusuke Oda. 09:10–10:00. Keynote 1 Jacob Devlin. 10:00–10:30. Shared Task Overview. 10:30–11:00. Coffee Break. 11:00–11:30. Marian: Fast Neural Machine Translation in C++. 11:30–12:20. Keynote 2 Rico Sennrich. 12:20–13:20. Lunch Break. 13:20–13:50. Best Paper Session. 13:50–14:40. Keynote 3 Jason Weston. 14:40–15:30. Keynote 4 Yulia Tsvetkov. 15:30–16:00. Coffee Break. ix.

(10) July 20, 2018 (continued) 16:00–17:30. Poster Session A Shared Attention Mechanism for Interpretation of Neural Automatic Post-Editing Systems Inigo Jauregi Unanue, Ehsan Zare Borzeshi and Massimo Piccardi Iterative Back-Translation for Neural Machine Translation Vu Cong Duy Hoang, Philipp Koehn, Gholamreza Haffari and Trevor Cohn Inducing Grammars with and for Neural Machine Translation Yonatan Bisk and Ke Tran Regularized Training Objective for Continued Training for Domain Adaptation in Neural Machine Translation Huda Khayrallah, Brian Thompson, Kevin Duh and Philipp Koehn Controllable Abstractive Summarization Angela Fan, David Grangier and Michael Auli Enhancement of Encoder and Attention Using Target Monolingual Corpora in Neural Machine Translation Kenji Imamura, Atsushi Fujita and Eiichiro Sumita Document-Level Adaptation for Neural Machine Translation Sachith Sri Ram Kothur, Rebecca Knowles and Philipp Koehn On the Impact of Various Types of Noise on Neural Machine Translation Huda Khayrallah and Philipp Koehn Bi-Directional Neural Machine Translation with Synthetic Parallel Data Xing Niu, Michael Denkowski and Marine Carpuat Multi-Source Neural Machine Translation with Missing Data Yuta Nishimura, Katsuhito Sudoh, Graham Neubig and Satoshi Nakamura Towards one-shot learning for rare-word translation with external experts Ngoc-Quan Pham, Jan Niehues and Alexander Waibel. x.

(11) July 20, 2018 (continued) NICT Self-Training Approach to Neural Machine Translation at NMT-2018 Kenji Imamura and Eiichiro Sumita Fast Neural Machine Translation Implementation Hieu Hoang, Tomasz Dwojak, Rihards Krislauks, Daniel Torregrosa and Kenneth Heafield OpenNMT System Description for WNMT 2018: 800 words/sec on a single-core CPU Jean Senellart, Dakun Zhang, Bo WANG, Guillaume KLEIN, Jean-Pierre Ramatchandirin, Josep Crego and Alexander Rush Marian: Cost-effective High-Quality Neural Machine Translation in C++ Marcin Junczys-Dowmunt, Kenneth Heafield, Hieu Hoang, Roman Grundkiewicz and Anthony Aue On Individual Neurons in Neural Machine Translation D. Anthony Bau, Yonatan Belinkov, Hassan Sajjad, Nadir Durrani, Fahim Dalvi and James Glass Parameter Sharing Strategies in Neural Machine Translation Sébastien Jean, Stanislas Lauly and Kyunghyun Cho Modeling Latent Sentence Structure in Neural Machine Translation Joost Bastings, Wilker Aziz, Ivan Titov and Khalil Simaan Extreme Adaptation for Personalized Neural Machine Translation Paul Michel and Graham Neubig Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks Diego Marcheggiani, Joost Bastings and Ivan Titov 17:30–17:40. Closing Remarks. xi.

(12)

(13)

New documents

As shown in Figure 1 and Table 2, we have identified and sequenced 29 novel clones that do not match with other known expressed sequences but do match with fly genomic sequence

Standard deviation and coefficient of variation of blood glucose for comparison of glycemic control with intermittent measurement, two groups should be studied – both groups should have

To explore the dysregulated gene interactions in primary monocytes of XLA patients, the interaction networks were generated for DE protein-coding genes which were significantly enriched

Another possible null hypothesis would be 1 to chose a method for identifying tags, as we did with UniGene [7], and 2 to estimate the relevance of the possible extra bases according to

Cr-EDTA, glomerular filtration rate measured with infusion clearance of Cr-EDTA; CG, estimated glomerular filtration rate with the Cockcroft Gault equation using actual, preoperative

Liver samples were harvested from rats inoculated with normal saline and treated with water a, inoculated with normal saline and treated with HPE-XA-08 c, inoculated with TAA 200 mg/kg

anti-fXa, anti-activated factor X activity; aPCC, activated prothrombinase complex concentrate; aPTT, activated partial thromboplastin time; AUC, area under the curve; bid, twice daily;

Further investigation using SignalP v3.0 Table 3 predicted the presence of cleavage sites for bacterial signal peptidases, 10 commonly expressed proteins i.e PotF, GltI, ArgT, Cbp,

From this research, the composition percentage of carbon dioxide and carbon monoxidefrom the combustion of direct injection of internal combustion engine can be determined according to

The article also attempts to at least partially explain the practices, querying for instance whether the disproportionately high amounts of citations to Pre-Trial Chamber records are a