JiaXu

Jia Xu

Assistant Professor
Graduate Center & Hunter College
City University of New York (CUNY)
(On leave from ICT/CAS)
CV

Contact Info
Department of Computer Science
Hunter College, North Building - Rm. 1043
695 Park Avenue, New York, NY 10065
jia.xu@hunter.cuny.edu

news

  • Congrats to MSc Geliang Chen! 12/2016
  • Paper accepted at AAAI 11/2016
  • Starting at CUNY Fall'16! 9/2016
  • Received joint NSFC grant 8/2016
  • Invited talk at Rutgers 8/2016
  • MIT Deep Learning Workshop 6/2016
  • Paper accepted at ICML 5/2016
  • Invited talk at Google Research NYC 12/2015
  • Invited talk at Columbia 11/2015
  • Invited talk at U. of Washington 11/2015
  • Invited talk at USC 11/2015
    Won 4th place (overall) and 1st (academic) at the NIST MT Competition 6/2015

I am creating methods for competitive machine translation systems. These methods often push things beyond the current state-of-the-art. To achieve this, I am devising general machine learning methods, study their empirical and theoretical limitations, and introduce techniques in ensemble learning, subsampling methods, and bringing geometric techniques in the study of structured prediction.

I am an assistant professor at the Graduate Center and Hunter college at the City University of New York (CUNY). Prior to coming to CUNY I was an associate professor at the Institute of Computing Technology, Chinese Academy of Sciences (Beijing). Before that I was an assistant professor for three years and as the chair of the elite CS undergraduate program, IIIS (Yao-class) at Tsinghua University (Beijing). Before Tsinghua I was a project leader and senior researcher in the language technology group at DFKI, Germany. As a graduate student I was supervised by Hermann Ney at RWTH Aachen with occasional (each a few-months long) and very fruitful visits to the speech group in IBM Watson and the NLP group in Microsoft Research (MSR) Redmond. My current research interests are in Machine Learning with a focus towards highly competitive machine translation systems. Lately, I have developed an interest and devise techniques that explore the underlying Metric and Geometric properties of machine translation systems. I am publishing in mainstream venues in computational linguistics and machine learning (e.g. AAAI, ICML, ACL) and often times I am leading teams that win first (or one of the first) position in machine translation competitions.

On the Efficient Online Model Adaptation by Incremental Simplex Tableau

Zhixian Lei, Xuehan Ye, Yongcai Wang, Deying Li, Jia Xu


On the Power and Limits of Distance-Based Learning

P. A. Papakonstantinou, J. Xu, G. Yang (authors alphabetically ordered)


System Description of ICT/Dublin NIST-15 Machine Translation System

J. Xu


Bagging by Design (On the Sub-optimality of Bagging)

P. A. Papakonstantinou, J. Xu, Z. Cao (first two authors alphabetically ordered)


Query Lattice for Translation Retrieval

M. Dong, Y. Cheng, Y. Liu, J. Xu, M. Sun


Salient Object Detection in Image Sequences via Spatial-Temporal Cue

C. Gan, Z. Qin, J. Xu and T. Wan


Enhancing Chinese Word Segmentation Using Unlabeled Data

W. Sun and J. Xu

July 2011, Proceedings of the Conference on Empirical Methods in Natural Language Processing


Generating Virtual Parallel Corpus: A Compatibility Centric Method

J. Xu and W.Sun

September 2011, Proceedings of the Machine Translation Summit


Improving Machine Translation Performance Using Comparable Corpora

A. Eisele and J. Xu

May 2010, Proceedings of the LREC Workshop on Building and Using Comparable Corpora


Synchronous Learning of Chinese Word Segmentation and Word Alignment

J. Xu, J.Gao, K.Toutanova and H. Ney

2011, Handbook of Natural Language Processing and Machine Translation (Book chapter)


Further Experiments with Shallow Hybrid MT Systems

C. Federmann, A. Eisele, Y. Chen, S. Hunsicker, J. Xu and H. Uszkoreit

July 2010, Proceedings of the ACL Workshop on Statistical Machine Translation


Bayesian Semi-Supervised Chinese Word Segmentation for Statistical Machine Translation

J. Xu, J. Gao, K. Toutanova and H. Ney

August 2008, Proceedings of the 22nd International Conference on Computational Linguistics


Phrase Table Training for Precision and Recall: What Makes a Good Phrase and a Good Phrase Pair?

Y.Deng, J. Xu and Y. Gao

June 2008, Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics


Domain Dependent Machine Translation

J. Xu, Y. Deng, Y. Gao and H. Ney

September 2007, Proceedings of the Machine Translation Summit XI


Partitioning Parallel Documents Using Binary Segmentation

J. Xu, R. Zens and H. Ney

June 2006, Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL): Proceedings of the Workshop on Statistical Machine Translation


Error Analysis of Statistical Machine Translation Output

D.Vilar, J. Xu, L. F. D'Haro and H. Ney

May 2006, Proceedings of the 5th International Conference on Language Resources and Evaluation (LREC)


Integrated Chinese Word Segmentation in Statistical Machine Translation

J. Xu, E. Matusov, R. Zens and H. Ney

October 2005, Proceedings of the International Workshop on Spoken Language Translation (IWSLT)


The RWTH Phrase-based Statistical Machine Translation System

R. Zens, O. Bender, S. Hasan, S. Khadivi, E. Matusov, J. Xu, Y. Zhang, and H. Ney

May 2005, Proceedings of the International Workshop on Spoken Language Translation (IWSLT)


Sentence Segmentation Using IBM Word Alignment Model 1

J. Xu, R. Zens and H. Ney

May 2005, Proceedings of the 10th Annual Conference of the European Association for Machine Translation (EAMT 2005)


Do We Need Chinese Word Segmentation for Statistical Machine Translation?

J. Xu, R. Zens and H. Ney

July 2004, Proceedings of the 10th Annual Conference of the European Association for Machine Translation (EAMT 2005)


Sequence Segmentation of Statistical Machine Translation

J. Xu

September 2010, Dissertation of Ph.D. of Computer Science


A Computational Model for Sound Processing in the Human Auditory System

J. Xu

November 2002, Dissertation of Master of Computer Science


I led ICT-DCU team achieving Rank 4 (among all participants) and Rank 1 (among academic affiliations) in NIST MT competition (on Chinese-English task, 2015): the best ranking record of ICT and of DCU in history of NIST participation.

What is the NIST MT competition?

Competition Representing Rank Notes
NIST 2015 ICT-DCU 1st and 4th 1st (academic inst.) and 4th (overall)
team leader and main contributor
WMT 2011 DFKI 1st team leader -- Official result
NIST 2008 MSR 1st intern at MSR -- Official result
NIST 2006 RWTH-Aachen 4th Official result
NIST 2005 RWTH-Aachen 4th Official result
NIST 2004 RWTH-Aachen 2nd Official result
GALE 2008 RWTH-Aachen 2nd ranked second in NightInGale
GALE 2007 RWTH-Aachen 2nd ranked second in NightInGale
GALE 2006 RWTH-Aachen 2nd ranked second in NightInGale
TC-Star 2006 RWTH-Aachen 1st
TC-Star 2005 RWTH-Aachen 1st
TC-Star 2004 RWTH-Aachen 1st
Source Amount Role Duration Title Grant No.
NSFC (NSF-China) grant 660,000 RMB (100,000 USD) Co-PI 2017--2019 Key Problems for Tightly-coupled, Multi-signal Fusion based Simultaneously Locating and Mapping --
ICT-CAS grant (Innovation subjects) 500,000 RMB (83,000 USD) Principal investigator 2015--2017 Ensemble learning in machine translation 20156020
KLIIP-ICT-CAS grant 200,000 RMB (33,000 USD) Principal investigator 2015 -- 2016 Novel machine learning methods 20156020
NSFC grant 660,000 RMB (100,000 USD) Co-PI 2014-2017 New approaches to the limits of efficient propositional reasoning: algorithms, approximations and foundations 20131351464
IIIS-Tsinghua grant 150,000 RMB (25,000 USD) Principal investigator 2012-2015 Machine learning and machine translation NA