Fluency Constraints for Minimum Bayes-Risk Decoding of Statistical Machine Translation Lattices

Download: Slides.

Fluency Constraints for Minimum Bayes-Risk Decoding of Statistical Machine Translation Lattices” by G. Blackwood, A. de Gispert, and W. Byrne. In Proceedings of the International Conference on Computational Linguistics (COLING), 2010, pp. 71-79 (9 pages).

Abstract

A novel and robust approach to incorporating natural language generation into statistical machine tr anslation is developed within a minimum Bayes-risk decoding framework. Segmentation of translation l attices is guided by confidence measures over the maximum likelihood translation hypothesis in order to focus on regions with potential translation errors. Modeling techniques intended to improve flue ncy in low confidence regions are introduced so as to improve overall translation fluency.

Download: Slides.

BibTeX entry:

@inproceedings{fluencycoling2010,
   author = {G. Blackwood and A. de Gispert and W. Byrne},
   title = {Fluency Constraints for Minimum {B}ayes-Risk Decoding of
	Statistical Machine Translation Lattices},
   booktitle = {Proceedings of the International Conference on
	Computational Linguistics (COLING)},
   pages = {71--79 (9 pages)},
   year = {2010},
   url = {http://www.aclweb.org/anthology/C/C10/C10-1009.pdf}
}

Back to Bill Byrne publications.