Biography
I am a graduate student working with Michael Collins at MIT CSAIL, and I have previously received a B.S. and M.Eng. in EECS at MIT. My broad research interests are discriminative methods and the use of unlabeled data in structured NLP tasks.
I am currently working on methods for integrating unlabeled data into dependency parsing (both projective and nonprojective), using the framework of Ando and Zhang (ACL 2005). I am also interested in exploring more specific NLP tasks, such as resolving ambiguous coordinations.
Here's a page that attempts to track activity on the Rosetta servers.
Publications
X. Carreras, M. Collins, and T. Koo. TAG, Dynamic Programming, and the Perceptron for Efficient, Feature-rich Parsing. Proceedings of CoNLL, 2008. Best paper award.
T. Koo, X. Carreras, and M. Collins. Simple Semi-supervised Dependency Parsing. Proceedings of ACL, 2008.
M. Collins, A. Globerson, T. Koo, X. Carreras, and Peter Bartlett. Exponentiated Gradient Algorithms for Conditional Random Fields and Max-Margin Markov Networks. Journal of Machine Learning Research 9(Aug):1775–1822, 2008.
T. Koo, A. Globerson, X. Carreras, and M. Collins. Structured Predictcion Models via the Matrix-Tree Theorem. Proceedings of EMNLP, 2007.
A. Globerson, T. Koo, X. Carreras, and M. Collins. Exponentiated Gradient Algorithms for Log-Linear Structured Prediction. Proceedings of ICML, 2007.
T. Koo and M. Collins. Hidden-Variable Models for Discriminative Reranking. Proceedings of EMNLP, 2005.
M. Collins and T. Koo. Discriminative Reranking for Natural Language Parsing. Computational Linguistics 31(1):25-69, 2005.
T. Koo and M. Collins. Parse Reranking with WordNet Using a Hidden-Variable Model. Master's Thesis, 2004.
J. Hajič, M. Čmejrek, B. Dorr, Y. Ding, J. Eisner, D. Gildea, T. Koo, K. Parton, G. Penn, D. Radev, and O. Rambow. Natural Language Generation in the Context of Machine Translation; Section 3.3, Chapter 6 (with Jan Hajic). NLP WS-02 Final Report, 2002.
Talks
Machine Learning & Friends talk given at UMass Amherst.