Learning Tractable Word Alignment Models with Complex Constraints

Simple item page
dc.contributor.authorGanchev, Kuzman
dc.contributor.authorGraça, João V.
dc.contributor.authorTaskar, Ben
dc.date2023-05-17T06:02:46.000
dc.date.accessioned2023-05-22T19:19:48Z
dc.date.available2023-05-22T19:19:48Z
dc.date.issued2010-03-10
dc.date.submitted2011-02-01T10:26:29-08:00
dc.description.abstractWord-level alignment of bilingual text is a critical resource for a growing variety of tasks. Probabilistic models for word alignment present a fundamental trade-off between richness of captured constraints and correlations versus efficiency and tractability of inference. In this article, we use the Posterior Regularization framework (Graça, Ganchev, and Taskar 2007) to incorporate complex constraints into probabilistic models during learning without changing the efficiency of the underlying model. We focus on the simple and tractable hidden Markov model, and present an efficient learning algorithm for incorporating approximate bijectivity and symmetry constraints. Models estimated with these constraints produce a significant boost in performance as measured by both precision and recall of manually annotated alignments for six language pairs. We also report experiments on two different tasks where word alignments are required: phrase-based machine translation and syntax transfer, and show promising improvements over standard methods.
dc.description.commentsSuggested Citation: J. Graça, K. Ganchev and B. Taskar. (2010). "Learning TractableWord AlignmentModels with Complex Constraints." Computational Linguistics. Vol. 36(3). p. 481-504. © 2010 MIT Press http://www.mitpressjournals.org/loi/coli
dc.identifier.urihttps://repository.upenn.edu/handle/20.500.14332/34820
dc.legacy.articleid1068
dc.legacy.fieldstrue
dc.legacy.fulltexturlhttps://repository.upenn.edu/cgi/viewcontent.cgi?article=1068&context=grasp_papers&unstamped=1
dc.source.issue66
dc.source.journalLab Papers (GRASP)
dc.source.peerreviewedtrue
dc.source.statuspublished
dc.subject.otherEngineering
dc.titleLearning Tractable Word Alignment Models with Complex Constraints
dc.typeArticle
digcom.identifiergrasp_papers/66
digcom.identifier.contextkey1756674
digcom.identifier.submissionpathgrasp_papers/66
digcom.typearticle
dspace.entity.typePublication
relation.isAuthorOfPublication5b360bfb-5497-43dc-9a15-0236987ccc59
relation.isAuthorOfPublication5b360bfb-5497-43dc-9a15-0236987ccc59
relation.isAuthorOfPublication48084f74-55a3-43da-96d7-8a01c512b3b9
relation.isAuthorOfPublication.latestForDiscovery5b360bfb-5497-43dc-9a15-0236987ccc59
upenn.schoolDepartmentCenterLab Papers (GRASP)
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Learning_Tractable_Word_Alignment.pdf
Size:
684.46 KB
Format:
Adobe Portable Document Format
Download
Collection
Articles