uu.seUppsala University Publications
Change search
ReferencesLink to record
Permanent link

Direct link
To Cache or not to Cache?: Experiments with Adaptive Models in Statistical Machine Translation
Uppsala University, Disciplinary Domain of Humanities and Social Sciences, Faculty of Languages, Department of Linguistics and Philology. (Computational Linguistics)
2010 (English)In: Proceedings of the ACL 2010 Joint Fifth Workshop on Statistical Machine Translation and Metrics MATR, Stroudsburg, PA, USA: Association for Computational Linguistics (ACL) , 2010Conference paper (Refereed)
Abstract [en]

We report results of our submissions to the WMT 2010 shared translation task in which we applied a system that includes adaptive language and translation models. Adaptation is implemented using exponentially decaying caches storing previous translations as the history for new predictions. Evidence from the cache is then mixed with the global background model. The main problem in this setup is error propagation and our submissions essentially failed to improve over the competitive baseline. There are slight improvements in lexical choice but the global performance decreases in terms of BLEU scores.

Place, publisher, year, edition, pages
Stroudsburg, PA, USA: Association for Computational Linguistics (ACL) , 2010.
Keyword [en]
machine translation, SMT
National Category
Language Technology (Computational Linguistics) Language Technology (Computational Linguistics)
Research subject
Computational Linguistics
URN: urn:nbn:se:uu:diva-126391ISBN: 978-1-932432-71-8OAI: oai:DiVA.org:uu-126391DiVA: diva2:323712
ACL 2010, Uppsala, Sweden, July 11–16, 2010
Available from: 2010-06-11 Created: 2010-06-11 Last updated: 2011-07-11Bibliographically approved

Open Access in DiVA

No full text

Other links

Electronic full text
By organisation
Department of Linguistics and Philology
Language Technology (Computational Linguistics)Language Technology (Computational Linguistics)

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 161 hits
ReferencesLink to record
Permanent link

Direct link