Mixture-Model Adaptation for SMT

Download	View accepted manuscript: Mixture-Model Adaptation for SMT (PDF, 271 KiB)
Author	Search for: Foster, George; Search for: Kuhn, Roland
Format	Text, Article
Conference	The ACL 2007 Workshop on Statistical Machine Translation (WMT-07), June 23 2007, Prague, Czech Republic
Abstract	We describe a mixture-model approach to adapting a Statistical Machine Translation system for new domains, using weights that depend on text distances to mixture components. We investigate a number of variants on this approach, including cross-domain versus dynamic adaptation; linear versus loglinear mixtures; language and translation model adaptation; different methods of assigning weights; and granularity of the source unit being adapted to. The best methods achieve gains of approximately one BLEU percentage point over a state-of-the art non-adapted baseline system.
Date published	2007
In	StatMT'07: Proceedings of the Second Workshop on Statistical Machine Translation: 128–135.
Language	English
NRC number	NRCC 49351
NPARC number	8914141
Export citation	Export as RIS
Report a correction	Report a correction (opens in a new tab)
Record identifier	22ebfcf1-86e6-4af8-a8c3-ea3d51ec5b75
Record created	2009-04-22
Record modified	2020-08-12

Page details