Paper: Faster MT Decoding Through Pervasive Laziness

ACL ID N09-2036
Title Faster MT Decoding Through Pervasive Laziness
Venue Human Language Technologies
Session Short Paper
Year 2009
Authors

Syntax-based MT systems have proven effective—the models are compelling and show good room for improvement. However, decoding involves a slow search. We present a new lazy-search method that obtains signifi- cant speedups over a strong baseline, with no loss in Bleu.