Multi-pivot translation by system combination

Gregor Leusch, Aurélien Max, Josep Maria Crego, Hermann Ney


Abstract
This paper describes a technique to exploit multiple pivot languages when using machine translation (MT) on language pairs with scarce bilingual resources, or where no translation system for a language pair is available. The principal idea is to generate intermediate translations in several pivot languages, translate them separately into the target language, and generate a consensus translation out of these using MT system combination techniques. Our technique can also be applied when a translation system for a language pair is available, but is limited in its translation accuracy because of scarce resources. Using statistical MT systems for the 11 different languages of Europarl, we show experimentally that a direct translation system can be replaced by this pivot approach without a loss in translation quality if about six pivot languages are available. Furthermore, we can already improve an existing MT system by adding two pivot systems to it. The maximum improvement was found to be 1.4% abs. in BLEU in our experiments for 8 or more pivot languages.
Anthology ID:
2010.iwslt-papers.12
Volume:
Proceedings of the 7th International Workshop on Spoken Language Translation: Papers
Month:
December 2-3
Year:
2010
Address:
Paris, France
Venue:
IWSLT
SIG:
SIGSLT
Publisher:
Note:
Pages:
299–306
Language:
URL:
https://aclanthology.org/2010.iwslt-papers.12
DOI:
Bibkey:
Cite (ACL):
Gregor Leusch, Aurélien Max, Josep Maria Crego, and Hermann Ney. 2010. Multi-pivot translation by system combination. In Proceedings of the 7th International Workshop on Spoken Language Translation: Papers, pages 299–306, Paris, France.
Cite (Informal):
Multi-pivot translation by system combination (Leusch et al., IWSLT 2010)
Copy Citation:
PDF:
https://aclanthology.org/2010.iwslt-papers.12.pdf