Call for papers
Workshop on Data-driven MT
Invited speaker: Hermann Ney, RWTH Aachen
Deadline for paper submissions: April 6, 2001
Deadline for notification of paper acceptance: April 27, 2001
Deadline for camera-ready papers: May 16, 2001
Workshop Date: July 7, 2001
Details on submissions listed below.
With the increased availability of online corpora, data-driven
approaches have become central to the NL community. A variety of
data-driven approaches have been used to help build Machine Translation
systems -- example-based, statistical MT, and other machine learning
approaches - and there are all sorts of possibilities for hybrid systems.
We wish to bring together proponents of as many techniques as possible to
engage in a discussion of which combinations will yield maximal success in
We propose to center the workshop on Data Driven MT, by which we
mean all approaches which develop algorithms and programs to exploit data
in the development of MT, primarily the use of large bilingual corpora
created by human translators, and serving as a source of training data for
MT systems. We are specifically interested in papers about
* statistical machine translation (modeling, training,
* machine-learning in translation
* example-based machine translation
* acquisition of multilingual training data
* evaluation of data driven methods (also with
* combination of various translation systems;
integration of classical rule-based and data driven approaches
* word/sentence alignment methods
An especially important question that we wish to address is which
techniques are best for each of the subparts of a complete MT system -
e.g. learning grammars, building lexicons, parsing input data,
determining transfer principles, generating target text, etc.
We will strongly encourage papers on systems which show
demonstrable progress over previously chosen methods, and which have been
integrated in an actual end-to-end system. Test results or demos will be
given strongest preference for participation.
Jessie Pinkham, Microsoft Research firstname.lastname@example.org
Kevin Knight USC/ISI; email@example.com <mailto:firstname.lastname@example.org
Web page http://www.isi.edu/~knight/
Franz Josef Och, RWTH Aachen; email@example.com
SUBMISSION FORMAT AND INSTRUCTIONS:
Electronic submissions only; send the postscript or pdf form of your
submission to: Deborah Coughlin firstname.lastname@example.org .
Submissions should follow the two-column format of ACL proceedings
and should not exceed eight (8) pages, including references. We
strongly recommend the use of ACL LaTeX style files or Microsoft
Word Style files tailored for this year's conference. They are
available from the ACL-2001 program committee Web-site at
As reviewing will be blind, a separate identification page must be
sent by email. The identification page should include the paper title,
authors' names, affiliations, and email addresses, up to 5 keywords
specifying the subject area, and a short summary (up to 5 lines).
The paper should not include the authors' names and affiliations.
This archive was generated by hypermail 2b29 : Wed Feb 21 2001 - 09:05:18 MET