translation phrase 0.0040114
phrase translation 0.0040114
language translation 0.00335388
sentence paraphrase 0.0032613700000000004
translation evaluation 0.002879138
machine translation 0.0028294419999999997
chinese translation 0.0026891529999999997
translation experiments 0.002678103
translation probability 0.0026757449999999998
translation rules 0.0026633779999999997
translation error 0.002659298
paraphrase lattice 0.0026522670000000003
same paraphrase 0.0026354380000000004
translation quality 0.0026307889999999997
translation group 0.002627596
speech translation 0.002615105
translation lei 0.002603571
translation takashi 0.002589924
translation hailong 0.002587842
translation narges 0.002570949
turkish translation 0.002548354
paraphrase pairs 0.002537669
paraphrase probability 0.002460335
paraphrase length 0.0024476790000000003
paraphrase lattices 0.002433928
paraphrase list 0.002394101
proper paraphrase 0.002373322
linguistics paraphrase 0.0023702420000000003
paraphrase lat 0.002361957
paraphrase pair 0.0023561130000000004
building paraphrase 0.002342329
erroneous paraphrase 0.002336509
neous paraphrase 0.0023320930000000004
contents paraphrase 0.0023320930000000004
translation 0.0023044
same phrase 0.002253448
original phrase 0.00223442
phrase table 0.002228575
phrase extraction 0.002158411
phrase pairs 0.0021556789999999998
paraphrase 0.00208899
language model 0.00207536
appropriate phrase 0.002068143
phrase lattices 0.002051938
hierarchical phrase 0.002045846
phrase list 0.002012111
key phrase 0.0019938819999999998
priate phrase 0.001970582
augmented phrase 0.001950388
smt model 0.0018792880000000002
source sentence 0.0017786500000000001
training data 0.0017499
phrase 0.001707
original sentence 0.0016998
input sentence 0.0016976930000000001
model training 0.0016935600000000002
model score 0.001670784
word segmentation 0.0016499919999999999
different phrases 0.001608095
word alignment 0.0015868079999999999
sentence length 0.001531069
paraphrased sentence 0.001525195
human language 0.001504289
output sentence 0.001496734
language models 0.001493519
word meaning 0.001478821
other smt 0.001465977
last sentence 0.001462254
source smt 0.001459678
length word 0.0014592289999999998
training corpus 0.001440659
data issue 0.001420423
model probability 0.001397225
word suffix 0.001382501
sparse data 0.0013798060000000001
german word 0.0013777029999999999
proper paraphrases 0.001376242
smt system 0.001373348
parallel corpus 0.001373057
improving word 0.001366942
data costanza 0.001365544
data yassine 0.001365544
better data 0.001354082
data robert 0.0013472240000000002
compound word 0.001344215
noisy data 0.001342508
selection model 0.001339717
data representation 0.0013391520000000001
sumita language 0.001333765
multimodal data 0.001326053
language processing 0.001325769
natural language 0.001324815
language texts 0.001314764
model proba 0.0013075530000000001
entropy language 0.001302415
normalized language 0.001302135
field model 0.0012846230000000001
guage model 0.001284353
structured model 0.0012776760000000002
standard smt 0.001259128
