topic model 0.00505665
topic distribution 0.00477698
topic models 0.004682976
topic modeling 0.004510885
topic information 0.00442295
sentence topic 0.004353371
topic distributions 0.004282603
ing topic 0.0042634380000000005
bilingual topic 0.004255325
local topic 0.0042048020000000005
specific topic 0.00419735
sparse topic 0.004145773
topic granularity 0.004117878
topic dis 0.004098585
effect topic 0.004093089
topic assignments 0.004070126
topic distri 0.0040611150000000006
side topic 0.004054383
topic biases 0.0040502820000000005
probable topic 0.00404835
topic sharing 0.004042757
topic 0.00381852
word translation 0.00273709
translation model 0.00234282
new word 0.0020624989999999998
many word 0.002030945
word pair 0.001971562
word pairs 0.001920089
word trans 0.00190681
latent topics 0.001881238
separate word 0.00188095
model probability 0.0018582870000000001
many topics 0.0018558049999999999
conditioned word 0.001855638
training data 0.0018491010000000001
document distribution 0.001691746
disparate topics 0.001683214
propriate topics 0.0016803319999999999
lda model 0.001669767
model score 0.001640752
lexical translation 0.001623032
model features 0.001606962
linear model 0.001592598
baseline model 0.001590016
translation table 0.001587324
model weights 0.001567726
machine translation 0.00156273
model description 0.001532348
tion model 0.001528936
english translation 0.001508022
training corpus 0.001497526
translation probabilities 0.001484183
bitam model 0.001480915
training sentence 0.0014682760000000001
topics 0.00145726
translation quality 0.001452437
translation task 0.001451243
chinese data 0.0014425940000000002
same document 0.001425494
test data 0.001411727
translation systems 0.001381411
translation rules 0.001371776
correct translation 0.001363791
parallel training 0.001358456
data set 0.001350876
nist training 0.0013086859999999999
nist data 0.001290937
large training 0.0012830139999999999
training corpora 0.001255886
model 0.00123813
test document 0.001229337
tuning data 0.001189939
data statistics 0.001183173
cific distribution 0.001183022
training focus 0.001155487
training bitext 0.001155487
lexical probability 0.0011384989999999998
words 0.00113165
document level 0.001117545
latent dirichlet 0.001106826
translation 0.00110469
ltm models 0.001087289
gtm models 0.001086184
corpus sentences 0.001066482
chinese sentence 0.0010617690000000002
document context 0.001033466
unsupervised modeling 0.001032539
test sentence 0.001030902
lexical weight 9.95064E-4
separate document 9.81836E-4
distribution 9.5846E-4
document delineations 9.56906E-4
dirichlet allocation 9.50585E-4
nist corpus 9.39362E-4
training 9.33425E-4
same notion 9.316000000000001E-4
test set 9.312509999999999E-4
other features 9.246510000000001E-4
sentence level 9.191100000000001E-4
dirichlet parame 9.06186E-4
