| | |
Summary: Efficient Extraction of Oracle-best Translations from Hypergraphs
Zhifei Li and Sanjeev Khudanpur
Center for Language and Speech Processing and Department of Computer Science
The Johns Hopkins University, Baltimore, MD 21218, USA
zhifei.work@gmail.com and khudanpur@jhu.edu
Abstract
Hypergraphs are used in several syntax-
inspired methods of machine translation to
compactly encode exponentially many trans-
lation hypotheses. The hypotheses closest to
given reference translations therefore cannot
be found via brute force, particularly for pop-
ular measures of closeness such as BLEU. We
develop a dynamic program for extracting the
so called oracle-best hypothesis from a hyper-
graph by viewing it as the problem of finding
the most likely hypothesis under an n-gram
language model trained from only the refer-
ence translations. We further identify and re-
move massive redundancies in the dynamic
|
