model translation 0.00518755
translation probability 0.0034283309999999997
model translations 0.003364733
language model 0.003316228
machine translation 0.003230107
output translation 0.0031824139999999997
translation output 0.0031824139999999997
reference translation 0.003077855
translation outputs 0.003005565
japanese translation 0.0030046
systems model 0.003000837
poor translation 0.002980731
right translation 0.002950824
memt model 0.002931492
translation literature 0.002931213
ticular translation 0.0029287239999999997
mixture model 0.002913652
confidence model 0.002873812
ibm model 0.002869313
based model 0.002840613
regression model 0.0027928560000000002
guage model 0.002788711
model tcpc 0.002777338
ther model 0.002743696
translation 0.0026858
model 0.00250175
training data 0.002272842
word length 0.002125814
data set 0.002030581
data sets 0.001887631
language models 0.001883648
train data 0.001868772
block word 0.0018579
japanese data 0.00185728
corresponding data 0.00183723
input data 0.001829129
right data 0.001803504
wrong data 0.001780481
alignment models 0.001678291
source sentence 0.001554386
memt models 0.0014989119999999998
test sentences 0.001472987
confidence models 0.001441232
other translations 0.001438979
target sentences 0.001418088
english sentence 0.001390629
regressive models 0.001387794
source sentences 0.001381773
bleu score 0.0013605890000000002
training alignment 0.001343483
svr models 0.00133292
models momentar 0.0013115099999999999
human translations 0.00131017
possible translations 0.001298807
sentence level 0.001274389
test set 0.001237955
different sources 0.0012335319999999999
bilingual sentences 0.001226239
dynamic language 0.001225998
sentence basis 0.001217946
different line 0.0012123939999999999
input sentence 0.001190395
multiple search 0.0011731609999999998
different engines 0.001157587
candidate translations 0.001142068
confidence score 0.001136986
parallel corpus 0.001133954
source speech 0.001121104
translations block 0.001111443
memt algorithm 0.001103464
corpus size 0.001097854
same set 0.001096274
test sets 0.001095005
hfa score 0.0010802350000000001
high score 0.001078029
other systems 0.001075083
voting language 0.001070838
models 0.00106917
search engines 0.001050481
possible approach 0.001044541
source text 0.0010374899999999999
reliable score 0.001037166
gual corpus 0.0010222080000000001
probability estimates 0.001016689
small set 0.001006189
test block 9.94314E-4
test col 9.88232E-4
nii test 9.88232E-4
target values 9.856320000000002E-4
same level 9.78816E-4
memt performance 9.60878E-4
memt system 9.5692E-4
new domain 9.473999999999999E-4
open source 9.39462E-4
memt systems 9.288289999999999E-4
statistical information 9.28352E-4
statistical machine 9.2381E-4
particular approach 9.23009E-4
such lms 9.21785E-4
high bleu 9.087699999999999E-4
