Improving the minimum Bayes’ risk combination of machine translation systems

Jesús González-Rubio, Francisco Casacuberta


Abstract
We investigate the problem of combining the outputs of different translation systems into a minimum Bayes’ risk consensus translation. We explore different risk formulations based on the BLEU score, and provide a dynamic programming decoding algorithm for each of them. In our experiments, these algorithms generated consensus translations with better risk, and more efficiently, than previous proposals.
Anthology ID:
2013.iwslt-papers.4
Volume:
Proceedings of the 10th International Workshop on Spoken Language Translation: Papers
Month:
December 5-6
Year:
2013
Address:
Heidelberg, Germany
Venue:
IWSLT
SIG:
SIGSLT
Publisher:
Note:
Pages:
Language:
URL:
https://aclanthology.org/2013.iwslt-papers.4
DOI:
Bibkey:
Cite (ACL):
Jesús González-Rubio and Francisco Casacuberta. 2013. Improving the minimum Bayes’ risk combination of machine translation systems. In Proceedings of the 10th International Workshop on Spoken Language Translation: Papers, Heidelberg, Germany.
Cite (Informal):
Improving the minimum Bayes’ risk combination of machine translation systems (González-Rubio & Casacuberta, IWSLT 2013)
Copy Citation:
PDF:
https://aclanthology.org/2013.iwslt-papers.4.pdf