reordering model 0.002644464
language model 0.002564965
translation model 0.002481926
phrase reordering 0.002091374
specific model 0.0020128290000000003
probabilistic model 0.001990457
discriminative model 0.00198473
cal model 0.0019642410000000003
model consis 0.001943101
word alignment 0.001821876
training data 0.001734442
model 0.00170581
reordering problem 0.001669426
word align 0.001654125
word penalty 0.001641618
phrase pair 0.001510222
source phrase 0.001495892
phrase orientation 0.0014958130000000001
phrase pairs 0.001485479
different feature 0.0014815969999999999
phrase boundary 0.0014761190000000001
reordering models 0.0014756740000000002
target words 0.001474917
dual problem 0.001452907
current phrase 0.001449703
such reordering 0.001442342
feature vector 0.0014349340000000001
target phrase 0.0014348770000000002
training set 0.001433807
jacent words 0.001432774
phrase position 0.00141477
phrase movements 0.001398104
language models 0.001396175
phrase movement 0.001392919
phrase move 0.0013907050000000001
training corpus 0.0013779399999999998
scale reordering 0.001335991
dual variables 0.00123187
optimization problem 0.001211938
feature vectors 0.0012105240000000001
lexicalized reordering 0.001202791
words 0.00119276
language pairs 0.001191914
orthogonal feature 0.001187011
feature rep 0.001171383
feature vec 0.001169274
joint feature 0.001169079
data shrinking 0.001165142
class problem 0.001161374
phrase 0.00115272
classification problem 0.00115174
machine translation 0.001142387
training example 0.001126446
positive learning 0.001118479
machine learning 0.00110426
eral language 0.001098098
prediction problem 0.0010936919999999998
different models 0.0010928399999999999
translation systems 0.001078882
objective function 0.001077659
parallel corpus 0.001076886
dual problems 0.0010749610000000001
common problem 0.001064858
likely translation 0.001049665
learning rate 0.001038997
translation quality 0.0010281959999999999
score function 0.001020562
fluent translation 0.0010198709999999999
translation ebest 0.001018243
additional features 0.001015017
different languages 0.00101012
iterative learning 0.001002272
dual coordinate 9.85296E-4
dual prob 9.849610000000001E-4
online learning 9.83401E-4
learning literature 9.82747E-4
class probability 9.816809999999999E-4
primal problem 9.73412E-4
movement problem 9.709709999999999E-4
problem lebanon 9.69121E-4
problem require 9.69121E-4
main features 9.66975E-4
adjacent phrases 9.64592E-4
function difference 9.54458E-4
pirical distribution 9.505429999999999E-4
reordering 9.38654E-4
feature 9.25777E-4
sparse features 9.22572E-4
gradient method 9.20143E-4
linguistic features 9.01786E-4
contextual features 8.90377E-4
linear classifier 8.88439E-4
empirical results 8.70838E-4
language 8.59155E-4
results com 8.58271E-4
descent method 8.338110000000001E-4
training 8.16383E-4
discriminative models 8.1594E-4
variables αik 7.92617E-4
rior probability 7.91304E-4
