James Clarke & Research

Revisiting Optimal Decoding for Machine Translation IBM Model 4

Sebastian Riedel and James Clarke. 2009. Revisiting Optimal Decoding for Machine Translation IBM Model 4. In Proceedings of the NAACL HLT 2009 Short Papers, pages 5–8. Boulder, Colorado.

Abstract

This paper revisits optimal decoding for statistical machine translation using IBM Model 4. We show that exact/optimal inference using Integer Linear Programming is more practical than previously suggested when used in conjunction with the Cutting-Plane Algorithm. In our experiments we see that exact inference can provide a gain of up to one BLEU point for sentences of length up to 30 tokens.

Bibtex

@inproceedings{Riedel:Clarke:09,
  author =       {Sebastian Riedel and James Clarke},
  title =        {Revisiting Optimal Decoding for Machine Translation
                  IBM Model~4},
  booktitle =    {Proceedings of the NAACL HLT 2009 Short Papers},
  pages =        {5--8},
  year =         {2009},
  address =      {Boulder, Colorado},
  URL =
                 {http://jamesclarke.net/media/papers/riedel-clarke-naacl09.pdf},
}