translation model 0.0023841369999999997
language model 0.0021740880000000002
word translation 0.0020996870000000003
alignment model 0.002026462
bayesian model 0.001842915
translation models 0.001772661
full model 0.001768159
model ing 0.001745323
word alignment 0.0017420120000000002
phrasal model 0.001732572
carlo model 0.00170513
model counts 0.001693839
based model 0.001680832
model parameters 0.001676611
model probabilities 0.0016725590000000001
distortion model 0.001658059
uniform model 0.001644861
ibm model 0.00163354
model param 0.001613144
model averaging 0.001613144
model exploits 0.001613144
word alignments 0.001569316
language models 0.0015626120000000001
output word 0.001467455
single word 0.001448751
word translations 0.0014480600000000001
word pairs 0.001422277
model 0.00140891
word pair 0.001385432
statistical word 0.001381538
translation process 0.001380811
word count 0.001376887
other language 0.001371645
machine translation 0.001356428
training data 0.001336178
full translation 0.001334476
phrasal translation 0.001298889
sampling time 0.0012869029999999998
translation evaluation 0.0012729759999999999
translation tasks 0.001266938
lexical translation 0.001263017
target language 0.001254108
translation systems 0.001236716
translation improvements 0.001232183
consistent translation 0.0012248559999999999
translation prob 0.001215327
training corpus 0.00120626
translation units 0.0012015819999999999
translation pipeline 0.00119256
translation sys 0.001182413
chine translation 0.001179214
phrase table 0.001171908
approach models 0.001166397
time complexity 0.001166301
target words 0.001123823
different types 0.0010932749999999999
training set 0.001083172
inference algorithm 0.00107131
individual models 0.001068798
other rule 0.001066408
frequency phrase 0.001066381
sampling algorithm 0.001063283
language pairs 0.0010629950000000002
phrase pairs 0.0010589990000000001
parallel data 0.001055754
time com 0.001042187
different modes 0.00103624
different colours 0.001031835
itg algorithm 0.001031314
average time 0.001025712
research topic 0.001025191
english data 0.0010192439999999999
training tree 0.001016104
sentence length 0.001013103
good phrase 0.001012062
source target 0.001009506
itg inference 0.001008692
phrase tables 0.0010004950000000001
alignment probability 0.001000484
variable sampling 9.81162E-4
translation 9.75227E-4
language modelling 9.704620000000001E-4
aggregate phrase 9.65592E-4
rate training 9.603740000000001E-4
bilingual data 9.55166E-4
other approaches 9.3166E-4
parallel corpus 9.25836E-4
training sets 9.15847E-4
bilingual training 9.09948E-4
inside inference 8.861100000000001E-4
full sampling 8.755659999999999E-4
true distribution 8.74716E-4
sample grammar 8.64729E-4
target sentences 8.642580000000001E-4
posterior distribution 8.57453E-4
itg tree 8.549720000000001E-4
training progress 8.497930000000001E-4
rule type 8.4741E-4
full itg 8.43597E-4
base distribution 8.426810000000001E-4
