language model 0.00391635
translation model 0.0038345899999999997
language word 0.00367572
model features 0.0030611839999999998
new model 0.002773794
target word 0.002757526
recurrent model 0.002731195
model score 0.002728872
network model 0.0026762089999999997
model experiments 0.002648105
language models 0.002626059
word vector 0.002625269
channel model 0.002621558
tion model 0.002603588
entropy model 0.002584627
joint model 0.002584319
trigram model 0.002560497
source language 0.002549758
model naı 0.0025487319999999997
model fea 0.002546418
translation models 0.002544299
guage model 0.002536352
model transla 0.00253231
model beam 0.002529423
unbounded model 0.0025278139999999998
nel model 0.002510727
model structure 0.002510346
word representations 0.00248798
word error 0.002481429
word vectors 0.002418715
lsa word 0.002409368
source words 0.002393598
next word 0.00238305
target language 0.002376126
input word 0.0023723579999999998
translation features 0.002357394
single word 0.002352863
word contexts 0.002352804
word counts 0.0023474669999999998
individual word 0.002329841
times word 0.002313872
word embeddings 0.002313139
foreign word 0.0023126789999999998
word encodings 0.002307532
word fea 0.002305788
bigram word 0.002305707
word vec 0.0022790889999999998
several language 0.0022693369999999997
model 0.00226919
target words 0.002219966
training data 0.00218491
baseline language 0.002167309
same translation 0.002139357
different translation 0.002130747
recurrent language 0.002109165
language pair 0.00208692
work language 0.002086822
machine translation 0.002065986
lexical translation 0.002059652
network language 0.0020541789999999997
many words 0.002021077
translation scores 0.002002503
new words 0.0019956039999999998
rnn language 0.001994748
language pairs 0.0019940039999999997
translation accuracy 0.00198542
translation tasks 0.001976804
joint language 0.001962289
various language 0.001957314
trigram language 0.001938467
language modeling 0.001932257
previous words 0.001924365
translation channel 0.001917768
conventional language 0.001907737
language projections 0.001906797
unbounded language 0.001905784
span language 0.001887534
translation hypothesis 0.001882964
reference translation 0.001882078
training corpus 0.001872351
translation hypotheses 0.001870401
translation direction 0.001863832
entire translation 0.00185635
translation modeling 0.001850497
input words 0.001834798
source sentence 0.001831892
translation directions 0.001816179
traditional translation 0.001810732
reference words 0.0018076779999999998
training time 0.001802682
foreign words 0.001775119
preceding words 0.001737228
language 0.00164716
translation 0.0015654
discriminative training 0.001560634
rate training 0.001545365
other features 0.001534958
ter training 0.0015342709999999998
standard phrase 0.001520932
bilingual data 0.001497385
