translation evaluation 0.00384678
evaluation metric 0.00380533
human evaluation 0.0037811299999999997
evaluation metrics 0.00360234
semantic evaluation 0.00329183
different evaluation 0.002940845
automatic evaluation 0.002897697
evaluation scores 0.002864574
evaluation reference 0.002834453
same evaluation 0.002800851
evaluation data 0.0027470899999999998
evaluation task 0.002604486
evaluation framework 0.002589565
standard evaluation 0.00257892
translation system 0.002574335
automatic metric 0.002568767
evaluation dataset 0.002548144
first evaluation 0.002538443
human scores 0.002511444
new evaluation 0.002455453
evaluation tasks 0.002452133
tion evaluation 0.002443177
comparative evaluation 0.002399192
conventional evaluation 0.002396271
automatic metrics 0.002365777
evaluation setting 0.002363358
detailed evaluation 0.002360338
evaluation methods 0.002356965
evaluation exercise 0.002355613
evaluation campaigns 0.0023544
evaluation campaign 0.0023500839999999997
machine translation 0.00233038
translation research 0.002285138
other metrics 0.002276878
output translation 0.002267576
translation output 0.002267576
translation quality 0.0022384650000000002
metric adequacy 0.002200383
single metric 0.002190513
translation outputs 0.002173731
human quality 0.002172815
translation tasks 0.002164653
human judgments 0.0021474800000000002
human evaluations 0.002118963
translation evalua 0.002092591
metric performance 0.002090335
translation challenge 0.00208429
metric definition 0.002073808
evaluation 0.00206713
standard metrics 0.002047
matic metric 0.002031389
human judges 0.002012588
human rankings 0.00200747
nist metrics 0.0019294899999999999
language model 0.001862628
ventional metrics 0.00181848
ard metrics 0.00181848
translation 0.00177965
source language 0.0017490539999999999
semantic framework 0.001747135
metric 0.0017382
semantic space 0.001701176
same correlation 0.001685818
semantic frame 0.001669498
sentence translations 0.0015655209999999998
metrics 0.00153521
semantic indexing 0.001533481
reference translations 0.001513831
training data 0.0015124119999999999
semantic embedding 0.001508192
output language 0.001473211
language quality 0.0014441
language models 0.001431111
sentence level 0.00136149
different dataset 0.0013547289999999998
space model 0.001353819
target language 0.001353349
system level 0.001337162
correlation coefficient 0.001318491
input sentence 0.001315426
parallel training 0.001310625
based language 0.001300881
fluency scores 0.0012936269999999999
language side 0.001291468
system output 0.0012826110000000001
available training 0.001281258
source meaning 0.0012784580000000001
model approach 0.001262394
correlation coefficients 0.001260757
different languages 0.0012566209999999999
word categories 0.001254188
correlation coef 0.001236121
correlation coeffi 0.001236121
vector model 0.001233032
different vector 0.001229404
individual sentence 0.001225836
model parameters 0.001218939
source vocabulary 0.0012170319999999998
model construction 0.001207294
system outputs 0.001188766
