translation model 0.0033394799999999997
rule set 0.002980107
translation forest 0.00297579
target translation 0.002919438
tree node 0.00282011
first rule 0.002777346
partial translation 0.002722051
translation process 0.00268434
rule matching 0.0026691989999999997
tree nodes 0.00266433
translation probabilities 0.002663351
translation hypothesis 0.002653469
machine translation 0.002611398
translation score 0.00258796
translation performance 0.002585044
quality translation 0.002566192
translation quality 0.002566192
translation scores 0.002559259
action rule 0.002551533
whole translation 0.002542027
stsg rule 0.002531089
tial translation 0.002525905
cfg rule 0.002521145
rule count 0.002502262
bidirectional translation 0.002484638
chine translation 0.002484638
translation mode 0.002484638
source tree 0.0024547939999999997
rightmost rule 0.002435993
applicable rule 0.0024335859999999997
grammar rules 0.002430063
rule corre 0.002424478
rule proba 0.002424478
different tree 0.0023666169999999997
same tree 0.002336296
decoding algorithm 0.0022916300000000002
string rules 0.002270432
many rules 0.002224565
other rules 0.002213944
translation 0.00220449
parse tree 0.002156219
form rules 0.002035972
only rules 0.002014145
stsg rules 0.002001749
cfg rules 0.001991805
tree structures 0.001975956
matched rules 0.001951065
cfg tree 0.001947425
tree substitution 0.001931838
applicable rules 0.001904246
plicable rules 0.001895315
same decoding 0.001891836
decoding time 0.001880785
chronous tree 0.001869903
btg tree 0.001864333
language model 0.001737866
source word 0.001715513
model probability 0.001692022
different source 0.001680211
source reordering 0.0016787310000000001
ing algorithm 0.0016690809999999998
root node 0.001663899
beam search 0.001646373
model context 0.001618398
rules 0.00161498
covered node 0.001598446
beam pruning 0.001594823
parsing algorithm 0.001586388
cube pruning 0.0015668589999999999
leaf node 0.0015663889999999998
model order 0.001560718
source sentence 0.0015473330000000001
target word 0.001546267
ghkm algorithm 0.001542238
end node 0.0015340719999999998
node labels 0.001530263
traditional decoding 0.001526194
phrase reordering 0.001509749
model evaluation 0.001501464
pruning method 0.0014941309999999999
cky decoding 0.001490725
source language 0.00148707
training set 0.001476918
source parse 0.001469813
decoding efficiency 0.0014552340000000001
algorithm groups 0.0014520269999999998
algorithm searches 0.0014490059999999999
algorithm constructs 0.00144684
word number 0.001443158
linear model 0.001438502
guage model 0.0014342579999999999
decoding speed 0.0014338340000000002
model probabil 0.001415373
avg decoding 0.0014140390000000002
source trees 0.0014084420000000002
search space 0.001404299
different beam 0.00138045
source words 0.001379292
same number 0.001377535
ered nodes 0.001376064
