We invite paper submissions to the ACL 2013 Workshop on Discourse in Machine Translation, to be held on August 9, 2013 in Sofia.
Over the past four years, there has been a resurgence of interest in machine translation from the perspective of discourse, and in discourse from the perspective of machine translation. This includes translating discourse phenomena that depend on more than just the context of the current sentence or of n-grams to the left and/or right; ensuring document-level consistency in the choice of lexical items or referring forms or in document style or register; and ensuring that source-language discourse relations between clauses and/or sentences are also realized in the target text.
The papers should focus on language processing techniques, whether theoretically-inspired or empirical, which address one or more of the discourse-level phenomena listed below, either in combination with MT, or from a cross-lingual perspective. We especially welcome papers that propose methods and software which enhance the capabilities of MT systems on discourse-level phenomena. Papers assessing the importance and origins of MT difficulties with discourse, including research and commercial MT systems, are suitable as well.
Topics
The proposed workshop solicits submissions more particularly focused on the following topics, but welcomes also submissions that link discourse studies with machine translation in some other way.
- discourse processing in support of MT, including:
o textual coherence, including anaphora, coreference, tense, aspect and modality
o textual cohesion, including lexical consistency
o discourse structure, including appropriate use of connectives and information structuring devices
o topic structure
o consistency in style and register
- MT techniques for obtaining document-level consistency and domain adaptability;
- MT techniques for structured documents;
- methods and algorithms to handle discourse-level phenomena in MT training and decoding;
- uses of MT in processing discourse-level phenomena;
- techniques for assessing the impact of discourse-level processing on MT quality;
- quantitative studies on the impact of discourse-level phenomena on current MT systems vs. discourse-aware ones.
Submission instructions
We solicit previously unpublished work, presented either as long or short papers, following the ACL 2013 formatting guidelines available at http://acl2013.org/site/call.html. Long papers should have at most 8 pages of content, not including references. Short papers are limited to 4 pages of content, not including references. There is no constraint on the length of the reference list. Both types of submissions should be anonymous, i.e. do not disclose in any way the identity of the author(s). Papers must be submitted using the START system at the URL that will be indicated later.
Website: http://www.idiap.ch/workshop/DiscoMT
Important dates
Submission deadline: April 26, 2013
Notification of acceptance: May 24, 2013
Final versions due: June 7, 2013
Workshop: August 9, 2013
Organizing Committee
- Bonnie Webber (University of Edinburgh), chair
- Ondrej Bojar (Charles University, Prague)
- Chris Callison-Burch (Johns Hopkins University)
- Marcello Federico (FBK-IRST, Trento)
- Pierre Isabelle (NCRC Canada)
- Katja Markert (University of Leeds), co-chair
- Andrei Popescu-Belis (Idiap Research Institute), co-chair
- Jörg Tiedemann (University of Uppsala), co-chair
Programme Committee
- Trevor Cohn (University of Sheffield)
- George Foster (NCRC Canada)
- Dan Gildea (University of Rochester)
- Liane Guillou (University of Edinburgh)
- Christian Hardmeier (University of Uppsala)
- Hitoshi Isahara (Toyohashi University of Technology)
- Philipp Koehn (University of Edinburgh)
- Thomas Meyer (Idiap Research Institute)
- Hwee Tou Ng (National University of Singapore)
- Michal Novak (Charles University Prague)
- Maja Popovic (DFKI)
- Jean Senellart (Systran)
- Lucia Specia (University of Sheffield)
- Sara Stymne (University of Uppsala)
- Gregor Thurmair (Linguatec GmbH)
- Sandrine Zufferey (Utrecht University)