Publication View

How Many Bits Are Needed To Store Probabilities for Phrase-Based Translation? (2006)

Abstract
State of the art in statistical machine translation is currently represented by phrasebased models, which typically incorporate a large number of probabilities of phrase-pairs and word n-grams. In this work, we investigate data compression methods for efficiently encoding n-gram and phrase-pair probabilities, that are usually encoded in 32-bit floating point numbers.

Publication details
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.61.6184
Source http://www.statmt.org/wmt06/proceedings/pdf/WMT13.pdf
Contributors CiteSeerX
Repository CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Type text
Language English
Relation 10.1.1.131.5458, 10.1.1.120.6608, 10.1.1.43.2981, 10.1.1.46.8369, 10.1.1.123.2204, 10.1.1.126.2845, 10.1.1.77.1633