training data 0.003118165
domain data 0.002838259
data smt 0.002630894
ing data 0.002601432
data selection 0.002597472
adaptation data 0.002562569
original data 0.002412403
data statistics 0.00240956
data setting 0.0023864
augmented data 0.0023820679999999998
data sizes 0.0023793039999999996
language model 0.00236013
model word 0.00223273
language models 0.002156688
model training 0.002028965
same word 0.0018626250000000001
neural language 0.001786344
language pairs 0.001684514
translation results 0.001639868
language pair 0.001602581
language mod 0.001567979
previous word 0.001545806
gram language 0.001545489
current word 0.001506325
training text 0.0014987120000000001
word contexts 0.001476682
training set 0.001473262
smt training 0.001465219
known word 0.001437747
word probabili 0.001434241
word coverage 0.001427504
domain sentences 0.0014272310000000002
same test 0.001370735
training time 0.00131803
language 0.00130741
lation model 0.001302709
unknown words 0.001291706
translation improvements 0.0012818109999999999
unaligned words 0.0012752549999999999
different lms 0.0012739449999999999
domain sentence 0.001273944
training times 0.0012594190000000001
training figure 0.0012589670000000002
domain text 0.001218806
reasonable training 0.001210606
chine translation 0.001207133
shows translation 0.001202959
different systems 0.001171038
source sentences 0.00112458
domain adaptation 0.001116988
smt system 0.001100467
selection method 0.001096879
guage models 0.001092766
reordering models 0.00108965
sentence set 0.001074622
similar sentences 0.001073995
phrase table 0.001062376
selection methods 0.00105784
modeling sentences 0.001053342
model 0.00105272
task system 0.001039941
adaptation methods 0.001022937
test sets 0.001017883
system bleu 0.00101578
words 0.00101501
training 9.76245E-4
translation 9.67661E-4
smt evaluation 9.60534E-4
evaluation table 9.59008E-4
sentence pairs 9.54709E-4
positive results 9.45241E-4
single system 9.1833E-4
probability value 9.14794E-4
effective approach 9.13234E-4
table adaptation 9.080970000000001E-4
other smoothing 9.07829E-4
neural networks 8.98469E-4
first work 8.97061E-4
corpus 8.75845E-4
sentence pair 8.72776E-4
neural lms 8.71548E-4
sentence position 8.688700000000001E-4
vector representations 8.614460000000001E-4
multiple smt 8.59264E-4
alldata system 8.493439999999999E-4
models 8.49278E-4
probability estimates 8.48197E-4
smooth probability 8.396E-4
advanced methods 8.38453E-4
raw probability 8.314E-4
phrase boundaries 8.11717E-4
smt train 8.08107E-4
validation set 7.90859E-4
state vector 7.90464E-4
smt systems 7.78681E-4
vector representation 7.77933E-4
continuous vector 7.74323E-4
tuning set 7.7015E-4
selection techniques 7.64856E-4
tinuous vector 7.581350000000001E-4
