Paper: Hierarchical System Combination for Machine Translation

ACL ID D07-1029
Title Hierarchical System Combination for Machine Translation
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2007

Given multiple translations of the same source sentence, how to combine them to produce a translation that is better than any single system output? We propose a hier- archical system combination framework for machine translation. This framework inte- grates multiple MT systems’ output at the word-, phrase- and sentence- levels. By boosting common word and phrase trans- lation pairs, pruning unused phrases, and exploring decoding paths adopted by other MT systems, this framework achieves bet- ter translation quality with much less re- decoding time. The full sentence translation hypotheses from multiple systems are addi- tionally selected based on N-gram language models trained on word/word-POS mixed stream, which further improves the transla- tion quality. We consistently observed sig- n...