translation model 0.0044710800000000005
phrase translation 0.00441463
translation table 0.003336896
phrases translation 0.003241991
translation probability 0.0032404870000000002
english translation 0.003160671
machine translation 0.003090383
translation pairs 0.003019982
reference translation 0.003001607
translation task 0.00300074
news translation 0.002989511
translation quality 0.002969309
supertagged translation 0.002953815
tic translation 0.002951595
model data 0.0029424200000000003
cal translation 0.0029246
translation candidates 0.002914107
language model 0.002775116
target phrase 0.002693689
translation 0.0026741
word alignment 0.002458465
lexical model 0.002454741
source phrase 0.002452588
baseline model 0.002409628
phrase table 0.002403326
lexical phrase 0.002398291
baseline phrase 0.002353178
same phrase 0.002295952
smt model 0.002278089
phrase probabilities 0.002174895
training data 0.00216219
markov model 0.0021606010000000003
supertag model 0.0021596180000000003
word sequences 0.002140977
word category 0.002140513
pbsmt model 0.002131053
linguistic word 0.00212947
phrase level 0.0021280599999999998
hierarchical phrase 0.002111694
word penalty 0.00210908
word projects 0.0021078390000000002
supertag phrase 0.002103168
bidirectional word 0.002092257
phrase pair 0.0020880869999999998
phrase pairs 0.0020864119999999997
word align 0.002076123
probabilistic model 0.0020754090000000003
model the 0.0020699100000000003
guage model 0.0020664660000000003
novel model 0.002064863
model statistics 0.002045582
model conditioning 0.002039104
supertagged phrase 0.002020245
phrase tables 0.002009584
phrase extraction 0.002009534
phrase cat 0.001973413
target language 0.001931295
target sentence 0.00180223
model 0.00179698
language models 0.001760532
phrase 0.00174053
system bleu 0.001724827
same data 0.001700862
data set 0.0016977989999999998
source language 0.001690194
baseline system 0.001675448
large data 0.00163415
ccg system 0.001604749
baseline language 0.001590784
training corpus 0.001586326
training set 0.001569109
source sentence 0.001561129
language phrases 0.0015460269999999997
smt system 0.001543909
other language 0.0015436399999999998
system output 0.001535351
parallel data 0.001527272
target phrases 0.00152105
small data 0.001510762
target sequence 0.001476012
language sentences 0.0014272389999999999
data sets 0.001423714
lexical syntactic 0.0014211010000000001
supertags system 0.001412842
system scores 0.001411573
other words 0.001409422
parallel training 0.0013985820000000002
pbsmt system 0.001396873
evaluation test 0.001389132
allel data 0.001386121
data sizes 0.001382658
arabic sentence 0.001374702
pair training 0.0013643070000000001
training corpora 0.001351919
supertagged system 0.001342515
markov language 0.0013417569999999998
supertag language 0.0013407739999999999
ltag system 0.0013364190000000001
language pairs 0.001324018
example system 0.001321205
