D11-1127 |
sections we report experiments in
|
Chinese-to-English translation
|
. For translation model training
|
D11-1080 |
We can observe the opposite for
|
Chinese-to-English translation
|
, where the slope is negative
|
D10-1054 |
large-scale translation tasks . On
|
Chinese-to-English translation
|
, the absolute improvements in
|
D08-1058 |
corresponding English terms . The
|
Chinese-to-English translation
|
performances of the two commercial
|
D13-1081 |
TER-BLEU over a strong baseline in
|
Chinese-to-English translation
|
. 1 Introduction Most of the
|
D13-1026 |
present our experiments on the NIST
|
Chinese-to-English translation
|
tasks . We first compare our
|
D09-1115 |
state-ofthe-art baseline system on
|
Chinese-to-English translation
|
test sets . 2 Background 2.1
|
D11-1081 |
work . 6.1 Data We focus on the
|
Chinese-to-English translation
|
task in this paper . The bilingual
|
D10-1053 |
the translations they yield . In
|
Chinese-to-English translation
|
, we find that rule extraction
|
D08-1011 |
proposed method achieved the best
|
Chinese-to-English translation
|
result in the constrained training
|
D08-1022 |
System Our experiments are on
|
Chinese-to-English translation
|
based on a tree-to-string system
|
D09-1114 |
( MT04 ) and 2005 ( MT05 ) for
|
Chinese-to-English translation
|
tasks . Both corpora provide
|
D09-1106 |
Preparation We evaluated our approach on
|
Chinese-to-English translation
|
. We used the FBIS corpus ( 6.9
|
D08-1010 |
Corpus We carry out experiments on
|
Chinese-to-English translation
|
. The training corpus is the
|
D11-1085 |
Experimental Results We report results on
|
Chinese-to-English translation
|
tasks using Joshua ( Li et al.
|
D12-1041 |
scale English-to-Japanese and
|
Chinese-to-English translation
|
tasks , and significant improvements
|
D09-1008 |
MT08 for Arabic - to-English and
|
Chinese-to-English translation
|
respectively . The context LM
|
D09-1114 |
may be useful to combination of
|
Chinese-to-English translation
|
, it is not applicable to most
|
D10-1054 |
tables , we observed that for
|
Chinese-to-English translation
|
, the most frequent " reordering
|
D12-1037 |
training into practice . In NIST
|
Chinese-to-English translation
|
tasks , our local training method
|