Fluency constraints for minimum Bayes-risk decoding of statistical machine translation lattices

  • Authors:
  • Graeme Blackwood;Adrià de Gispert;William Byrne

  • Affiliations:
  • Cambridge University;Cambridge University;Cambridge University

  • Venue:
  • COLING '10 Proceedings of the 23rd International Conference on Computational Linguistics
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

A novel and robust approach to improving statistical machine translation fluency is developed within a minimum Bayes-risk decoding framework. By segmenting translation lattices according to confidence measures over the maximum likelihood translation hypothesis we are able to focus on regions with potential translation errors. Hypothesis space constraints based on monolingual coverage are applied to the low confidence regions to improve overall translation fluency.