Combining Machine Translation Output with Open Source: The Carnegie Mellon Multi-Engine Machine Translation Scheme
K. Heafield, and A. Lavie. The Prague Bulletin of Mathematical Linguistics, (2010)
Abstract
The Carnegie Mellon multi-engine machine translation software merges output from several machine translation systems into a single improved translation. This improvement is significant: in the recent NIST MT09 evaluation, the combined Arabic-English output scored 5.22 BLEU points higher than the best individual system. Concurrent with this paper, we release the source code behind this result consisting of a recombining beam search decoder, the combination search space and features, and several accessories. Here we describe how the released software works and its use.
%0 Journal Article
%1 Heafield2010
%A Heafield, Kenneth
%A Lavie, Alon
%D 2010
%J The Prague Bulletin of Mathematical Linguistics
%K - LA Libre Software Traducci{\'{o}}n eng
%T Combining Machine Translation Output with Open Source: The Carnegie Mellon Multi-Engine Machine Translation Scheme
%U http://ufal.mff.cuni.cz/pbml/93/art-heafield-lavie.pdf
%X The Carnegie Mellon multi-engine machine translation software merges output from several machine translation systems into a single improved translation. This improvement is significant: in the recent NIST MT09 evaluation, the combined Arabic-English output scored 5.22 BLEU points higher than the best individual system. Concurrent with this paper, we release the source code behind this result consisting of a recombining beam search decoder, the combination search space and features, and several accessories. Here we describe how the released software works and its use.
@article{Heafield2010,
abstract = {The Carnegie Mellon multi-engine machine translation software merges output from several machine translation systems into a single improved translation. This improvement is significant: in the recent NIST MT09 evaluation, the combined Arabic-English output scored 5.22 BLEU points higher than the best individual system. Concurrent with this paper, we release the source code behind this result consisting of a recombining beam search decoder, the combination search space and features, and several accessories. Here we describe how the released software works and its use.},
added-at = {2015-12-01T11:33:23.000+0100},
author = {Heafield, Kenneth and Lavie, Alon},
biburl = {https://www.bibsonomy.org/bibtex/2d7fc788e1e2a3f4587fd9b2d7e8e3f38/sofiagruiz92},
interhash = {99864372ca352e086f6f1145db366802},
intrahash = {d7fc788e1e2a3f4587fd9b2d7e8e3f38},
journal = {The Prague Bulletin of Mathematical Linguistics},
keywords = {- LA Libre Software Traducci{\'{o}}n eng},
timestamp = {2015-12-01T11:33:23.000+0100},
title = {{Combining Machine Translation Output with Open Source: The Carnegie Mellon Multi-Engine Machine Translation Scheme}},
url = {http://ufal.mff.cuni.cz/pbml/93/art-heafield-lavie.pdf},
year = 2010
}