Abstract
System combination exploits differences between machine translation systems to form a combined translation from several system outputs. Core to this process are features that reward n-gram matches between a candidate combination and each system output. Systems differ in performance at the n-gram level despite similar overall scores. We therefore advocate a new feature formulation: for each system and each small n, a feature counts n-gram matches between the system and candidate. We show post-evaluation improvement of 6.67 BLEU over the best system on NIST MT09 Arabic-English test data. Compared to a baseline system combination scheme from WMT 2009, we show improvement in the range of 1 BLEU point.- Anthology ID:
- 2010.amta-papers.34
- Volume:
- Proceedings of the 9th Conference of the Association for Machine Translation in the Americas: Research Papers
- Month:
- October 31-November 4
- Year:
- 2010
- Address:
- Denver, Colorado, USA
- Venue:
- AMTA
- SIG:
- Publisher:
- Association for Machine Translation in the Americas
- Note:
- Pages:
- Language:
- URL:
- https://aclanthology.org/2010.amta-papers.34
- DOI:
- Cite (ACL):
- Kenneth Heafield and Alon Lavie. 2010. Voting on N-grams for Machine Translation System Combination. In Proceedings of the 9th Conference of the Association for Machine Translation in the Americas: Research Papers, Denver, Colorado, USA. Association for Machine Translation in the Americas.
- Cite (Informal):
- Voting on N-grams for Machine Translation System Combination (Heafield & Lavie, AMTA 2010)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2010.amta-papers.34.pdf