joint model 0.0034000000000000002
preposition model 0.002695933
model approach 0.002645451
learning model 0.002503649
model con 0.002428718
article model 0.0024095510000000002
individual model 0.002408723
baseline model 0.002404446
illinois model 0.002346231
discriminative model 0.002275663
task system 0.002272624
decoder model 0.002258356
expensive model 0.002255374
model han 0.002255374
systems joint 0.002233045
joint systems 0.002233045
joint inference 0.002202409
joint models 0.002173142
joint training 0.002117728
other word 0.002074358
correction system 0.002029541
model 0.00201994
joint approach 0.0020055710000000003
learning system 0.001985189
joint correction 0.001908121
article system 0.001891091
first word 0.001886264
joint learning 0.001863769
test data 0.0018600729999999998
training data 0.001848138
illinois system 0.001827771
system adjacent 0.001791879
data set 0.0017882459999999998
second word 0.001781636
joint components 0.001756914
target word 0.0017533610000000002
nois system 0.001738391
system below 0.001737548
word window 0.0017303980000000002
joint approaches 0.0017092399999999999
illinois joint 0.001706351
ith word 0.001699937
joint modeling 0.001689728
fusable word 0.001686594
joint methods 0.0016765460000000001
english data 0.0016704849999999998
sentence error 0.00162661
joint learn 0.001614943
error correction 0.001527697
task training 0.001508812
system 0.00150148
training test 0.001487271
esl data 0.00147547
gold data 0.001475435
inference approach 0.00144786
agreement error 0.001432514
other models 0.00141688
such models 0.0014079359999999998
grammatical error 0.001407494
learner data 0.001395274
other words 0.0013879489999999999
correction systems 0.001381046
joint 0.00138006
esl error 0.0013646359999999998
number error 0.001362349
error number 0.001362349
revised data 0.001356129
noun errors 0.001350192
inference problem 0.001347152
error analysis 0.0013428
error cor 0.0013356309999999999
second error 0.001330712
error rate 0.0013210729999999999
inference results 0.001301665
correction task 0.0012992049999999999
agreement errors 0.0012921
different part 0.001286116
input features 0.001281387
error type 0.0012812659999999999
many errors 0.001278733
learning models 0.001276791
noisy features 0.001274705
different evaluation 0.001269438
candidate words 0.001268309
practical error 0.0012567799999999999
error rates 0.00125062
several words 0.001248213
low error 0.001246152
individual systems 0.001241768
different types 0.001236878
error sparsity 0.0012357879999999998
rich features 0.0012349219999999998
esl errors 0.001224222
number errors 0.001221935
multiple errors 0.00122016
annotated errors 0.001218264
component inference 0.001202152
different evalu 0.00119679
individual models 0.001181865
set noun 0.001168746
