Unsupervised Word Alignment with Arbitrary Features

Unsupervised Word Alignment with Arbitrary Features

Cached

Download Links

by Chris Dyer , Jonathan Clark , Alon Lavie , Noah A. Smith

Citations:

3 - 0 self

Versions

Version 0

Version History

Metadata Version 0

Datum	Value	Source
TITLE	Unsupervised Word Alignment with Arbitrary Features	SVM HeaderParse 0.2
AUTHOR NAME	Chris Dyer	SVM HeaderParse 0.2
AUTHOR AFFIL	Language Technologies Institute; Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	Pittsburgh, PA 15213, USA	SVM HeaderParse 0.2
AUTHOR NAME	Jonathan Clark	SVM HeaderParse 0.2
AUTHOR AFFIL	Language Technologies Institute; Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	Pittsburgh, PA 15213, USA	SVM HeaderParse 0.2
AUTHOR NAME	Alon Lavie	SVM HeaderParse 0.2
AUTHOR AFFIL	Language Technologies Institute; Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	Pittsburgh, PA 15213, USA	SVM HeaderParse 0.2
AUTHOR NAME	Noah A. Smith	SVM HeaderParse 0.2
AUTHOR AFFIL	Language Technologies Institute; Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	Pittsburgh, PA 15213, USA	SVM HeaderParse 0.2
ABSTRACT	We introduce a discriminatively trained, globally normalized, log-linear variant of the lexical translation models proposed by Brown et al. (1993). In our model, arbitrary, nonindependent features may be freely incorporated, thereby overcoming the inherent limitation of generative models, which require that features be sensitive to the conditional independencies of the generative process. However, unlike previous work on discriminative modeling of word alignment (which also permits the use of arbitrary features), the parameters in our models are learned from unannotated parallel sentences, rather than from supervised word alignments. Using a variety of intrinsic and extrinsic measures, including translation performance, we show our model yields better alignments than generative baselines in a number of language pairs. 1	SVM HeaderParse 0.2
CITATIONS	40 found	ParsCit 1.0