Third Workshop on
Syntax and Structure in Statistical Translation
NAACL HLT 2009 Workshop
5 June 2009, Boulder, Colorado
The Third Workshop on Syntax and Structure in Statistical Translation (SSST-3) seeks to build on the foundations established in the first two SSST workshops, which brought together a large number of researchers working on diverse aspects of synchronous/transduction grammars (hereafter, S/TGs) in relation to statistical machine translation. Its program each year has comprised high-quality papers discussing current work spanning topics including: new grammatical models of translation; new learning methods for syntax-based models; using S/TGs for semantics and generation; syntax-based evaluation of machine translation; and formal properties of S/TGs. Presentations have led to productive and thought-provoking discussions, comparing and contrasting different approaches, and identifying the questions that are most pressing for future progress in this topic.
The need for structural mappings between languages is widely recognized in the fields of statistical machine translation and spoken language translation, and there is a growing consensus that these mappings are appropriately represented using a family of formalisms that includes synchronous/transduction grammars and their tree-transducer equivalents. To date, flat-structured models, such as the word-based IBM models of the early 1990s or the more recent phrase-based models, remain widely used. But tree-structured mappings arguably offer a much greater potential for learning valid generalizations about relationships between languages.
Within this area of research there is a rich diversity of approaches. There is active research ranging from formal properties of S/TGs to large-scale end-to-end systems. There are approaches that make heavy use of linguistic theory, and approaches that use little or none. There is theoretical work characterizing the expressiveness and complexity of particular formalisms, as well as empirical work assessing their modeling accuracy and descriptive adequacy across various language pairs. There is work being done to invent better translation models, and work to design better algorithms. Recent years have seen significant progress on all these fronts. In particular, systems based on these formalisms are now top contenders in MT evaluations.
We invite papers on:
* syntax-based / tree-structured statistical translation models and
* machine learning techniques for inducing structured translation models
* algorithms for training, decoding, and scoring with S/TGs
* empirical studies on adequacy and efficiency of formalisms
* studies on the usefulness of syntactic resources for translation
* formal properties of S/TGs
* scalability of structured translation methods to small or large data
* applications of S/TGs to related areas including:
o speech translation
o formal semantics and semantic parsing
o paraphrases and textual entailment
o information retrieval and extraction
For more information: http://www.cs.ust.hk/~dekai/ssst/
* Dekai WU (Hong Kong University of Science and Technology)
* David CHIANG (USC Information Sciences Institute)
* Srinivas BANGALORE (AT&T Research)
* Liang HUANG (Google)
* Kevin KNIGHT (USC Information Sciences Institute)
* Daniel MARCU (USC Information Sciences Institute)
* Yuji MATSUMOTO (Nara Institute of Science and Technology)
* Stefan RIEZLER (Google)
* Giorgio SATTA (Padua)
* Libin SHEN (BBN)
* Christoph TILLMANN (IBM)
* Stephan VOGEL (Carnegie Mellon University)
* Taro WATANABE (NTT)
* Yuk-Wah WONG (Google)
* Richard ZENS (Google)
Submission deadline: 6 Mar 2009 Notification to authors: 30 Mar 2009 Camera copy deadline: 12 Apr 2009
Papers will be accepted on or before 6 Mar 2009 in PDF or Postscript formats via the START system at https://www.softconf.com/naacl-hlt09/SSST3/. Submissions should follow the NAACL HLT 2009 length and formatting requirements for full papers of eight (8) pages of content with one (1) extra page for references, found at http://clear.colorado.edu/NAACLHLT2009/stylefiles.html.
Camera ready final versions will be accepted on or before 12 Apr 2009 in PDF or Postscript formats via the START system at https://www.softconf.com/naacl-hlt09/SSST3/. Papers should still follow the NAACL HLT 2009 length and formatting requirements for full papers, found at http://clear.colorado.edu/NAACLHLT2009/stylefiles.html.
Please send inquiries to ssst at cs.ust.hk.