Skip to Main Content
Table 14 

%BLEU on tune and test sets for UR→EN translation, comparing several settings for maximum dependency lengths in the decoder ( is for the source side and is for the target side). The upper table shows Moses BLEU scores for comparison. The lower table compares two max dependency length settings during tuning, and several for decoding on the test sets, showing both BLEU scores and average decoding times per sentence. See text for discussion.

graphic
 
graphic
 
Close Modal

or Create an Account

Close Modal
Close Modal