translation model 0.0052973099999999995
language translation 0.00464185
phrase translation 0.00435673
translation models 0.004074168
neural translation 0.004041558
space translation 0.0038300120000000003
translation system 0.003826695
machine translation 0.003797323
translation task 0.003735206
lexical translation 0.003700402
network translation 0.0037001440000000003
statistical translation 0.003689948
translation probabilities 0.003688723
translation iwslt 0.003640967
translation hypothesis 0.0036323270000000003
translation process 0.003605758
translation systems 0.0036017590000000004
tuple translation 0.0035794620000000003
translation direction 0.003560693
chine translation 0.003553533
trigram translation 0.0035442380000000003
alternative translation 0.003535587
gram translation 0.0035355670000000003
translation directions 0.0035339710000000003
translation mod 0.003533357
translation probabili 0.003529672
translation inputs 0.0035264790000000003
quality translation 0.003522499
translation quality 0.003522499
translation probabil 0.003522191
word language 0.0034741
language model 0.0034319599999999995
model training 0.00338895
translation 0.0032536
target word 0.0030603650000000002
word representation 0.002671015
word space 0.0026622620000000003
space model 0.002620122
word error 0.0026109740000000003
training data 0.00256296
word alignments 0.002542073
tion model 0.002512679
lexicon model 0.0024991889999999998
ibm model 0.0024981129999999997
first word 0.002494089
network model 0.002490254
last word 0.0024767120000000003
independent word 0.002459787
model framework 0.002392127
source words 0.00236856
alternative word 0.0023678370000000002
word choices 0.002365723
factored word 0.002363417
target language 0.002362765
word reorderings 0.0023556030000000004
dependent word 0.002352289
word bonus 0.0023470170000000003
word indices 0.002343968
trigram model 0.002334348
gram model 0.002325677
guage model 0.0023245279999999998
distorsion model 0.002322159
target words 0.002231275
language models 0.002208818
neural language 0.002176208
training sentence 0.002159343
target phrase 0.002077645
sentence words 0.002070863
model 0.00204371
bleu training 0.002030314
training corpus 0.002006998
words probability 0.001962259
approach words 0.0018654869999999999
ing data 0.001850536
same data 0.0018397180000000002
test data 0.001831834
target sentence 0.001788618
language pairs 0.001782295
neural target 0.001762473
phrase table 0.001745445
language sen 0.001741668
training text 0.0017368309999999999
language modeling 0.0017044059999999999
target probability 0.001680014
language mod 0.001668007
training algorithms 0.0016199159999999999
new phrase 0.0016098979999999998
training material 0.001604969
phrase pair 0.0016011189999999998
available data 0.001559564
bilingual sentence 0.001546414
eral words 0.001536227
ded words 0.001516749
data sparseness 0.001509411
development data 0.001509293
phrase pairs 0.001497175
opment data 0.001477025
neural bleu 0.001473032
source sen 0.001465218
layer output 0.001458088
