Graham Neubig's Research

Graham Neubig

Assistant Professor, Language Technology Institute, Carnegie Mellon University

My research is concerned with language and its role in human communication. In particular, my long-term research goal is to break down barriers in human-human or human-machine communication through the development of natural language processing (NLP) technologies. This includes the development of technology for machine translation, which helps break down barriers in communication for people who speak different languages, and natural language understanding, which helps computers understand and respond to human language. Within this overall goal of breaking down barriers to human communication, I have focused on several aspects of language that both make it interesting as a scientific subject, and hold potential for the construction of practical systems. Specific areas of interest include:

Machine Translation
Natural Language Processing (Text Analysis, Language Modeling, Parsing)
Machine Learning (Neural Networks, Un- and Semi-supervised Learning, Bayesian Methods)
Spoken Language Processing (Spoken Language Analysis, Spoken Dialog)
Software Language Processing (Conversion/Alignment between Code and Language)

Academic/Career History

9/2016-onward Carnegie Mellon University (CMU): Assistant Professor
4/2012-8/2016 Nara Institute of Science and Technology (NAIST): Assistant Professor
4/2010-3/2012 Kyoto University: Doctoral course in Intelligent Information Systems
4/2008-3/2010 Kyoto University: Master's course in Intelligent Information Systems
8/2006-3/2008 Hyogo Prefectural Government: Coordinator for International Relations
9/2005-7/2006 Tajima Agricultural High School: Assistant Language Teacher
8/2001-5/2005 University of Illinois, Urbana-Champaign: B.S. Computer Science

Awards

2017: 1st Workshop on Neural Machine Translation: Best Paper Award
2017: 2017 Meeting of the European Chapter of the Association for Computational Linguistics: Outstanding Paper Award
2016: 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP): Best Short Paper Award
2015: 2015 Meeting of the Association for Natural Language Processing: Outstanding Paper Award
2013: Journal of Information Processing Outstanding Paper Award
2012: Oriental COCOSDA Best Student Paper Award (Co-author)
2012: IPSJ SIG Recommended PhD Thesis
2012: Yamashita SIG Research Award
2012: JSAI Field Innovation Prize (as a member of the ANPI_NLP project)
2012: 2011 Meeting of the Association for Natural Language Processing: Outstanding Paper Award
2010: JSPS Research Fellowship for Young Scientists (DC1)

Visiting Researcher Experience

2015-2016 (Oct.-Jan.): CMU, Pittsburgh, USA. Focus on neural networks and natural language+code tasks. Host: Chris Dyer
2010-2012: NICT, Japan. Focus on unsupervised learning for machine translation. Host: Taro Watanabe
2011 (Jul.-Sep.): Google, Mountain View, USA. Research on machine translation. Host: David Talbot
2008 (Aug./Sep.): ATR, Japan. Researched language models for machine translation. Host: Andrew Finch

Activities

2016: Coling 2016, Machine Translation Area Chair
2016: ACL 2016, Machine Translation Area Chair
2015: Workshop on Asian Translation 2015, Organizer
2014, 2015: Kansai Machine Translation Study Group, Organizer
2013: Workshop on Language Processing and Crisis Information 2013, Organizer
2013, 2014: NLP2013 NLP2014, Tutorial Chair
2012, 2013: NLP2012 NLP2013, Organizer for the Theme Session on Language Information Processing in Times of Crisis
2011: NLP2011, Program Committee for the Workshop on the Relationship between Business, Universities, and Students in Natural Language Processing
Reviewer/PC: ACL, EACL, EMNLP, InterSpeech, ICASSP, JNLP (I will be happy to review for Open Access publications).

Presentations

Papers

Here is a list of a few of my current favorite papers:

Matthias Sperber, Graham Neubig, Jan Niehues, Alex Waibel.
Neural Lattice-to-Sequence Models for Uncertain Inputs (BibTex)
Conference on Empirical Methods in Natural Language Processing (EMNLP). Copenhagen, Denmark. September 2017 (To Appear).
Varun Gangal, Harsh Jhamtani, Graham Neubig, Eduard Hovy, Eric Nyberg.
Charmanteau: Character Embedding Models For Portmanteau Creation (BibTex)
Conference on Empirical Methods in Natural Language Processing (EMNLP). Copenhagen, Denmark. September 2017 (To Appear).
Chaitanya Malaviya, Graham Neubig, Patrick Littell.
Learning Language Representations for Typology Prediction (BibTex)
Conference on Empirical Methods in Natural Language Processing (EMNLP). Copenhagen, Denmark. September 2017 (To Appear).
Pengcheng Yin, Graham Neubig.
A Syntactic Neural Model for General-Purpose Code Generation (BibTex, Code/Data)
The 55th Annual Meeting of the Association for Computational Linguistics (ACL). Vancouver, Canada. July 2017.
Frederick Liu, Han Lu, Chieh Lo, Graham Neubig.
Learning Character-level Compositionality with Visual Features (BibTex, Code/Data)
The 55th Annual Meeting of the Association for Computational Linguistics (ACL). Vancouver, Canada. July 2017.

My most highly cited papers can be found on Google Scholar, and all of my publications can be found on my publications page.

Talks, etc.

Some others can be found on the tutorials page.

2015/09 Kyoto University Media Center Seminar: Machine Translation with Neural Nets (in Japanese)
2014/03 Kansai MT Study Group @ Hokkaido: I struggled all year to get an increase in human evaluation (in Japanese)
2014/03 ALAGIN Seminar @ Tokyo: 4 day machine translation tutorial (in Japanese), with Taro Watanabe, Katsuhito Sudoh, Akihiro Tamura, and Katsuhiko Hayashi. Slides about: Language Model, Translation Model, Alignment, Syntax-based Models, Data, Speech.
2013/01 ALAGIN Seminar @ Tokyo: 2 day machine translation tutorial (in Japanese).
2012/10 Kansai MT Study Group @ Kyoto: Experiments in English<->Japanese Tree-to-String Translation
2011/06 Microsoft Research Redmond: An unsupervised model for joint phrase alignment and extraction.
2011/05 Keihanna Speech Seminar @ NAIST: Non-parametric Bayesian methods (in Japanese)
2010/09 National Institute for the Japanese Language and Linguistics: Efficient corpus construction with active learning and pointwise prediction (in Japanese)
2010/08 NTT Communication Science Research Lab: Learning a language model from continuous speech