Skip to Main Content
Table 5: 
Consistency of the adversarial effect (or lack thereof) when retraining the models in the loop on the same data again, but with different random seeds. We report the mean and standard deviation (subscript) over 10 re-initialization runs.
OriginalRe-init.
ModelResourceEMF1EMF1
BiDAF DBiDAFdev 0.0 5.3 10.70.8 20.41.0 
BERT DBERTdev 0.0 4.9 19.71.0 30.11.2 
RoBERTa DRoBERTadev 0.0 6.1 15.70.9 25.81.2 
 
BiDAF DBiDAFtest 0.0 5.5 11.61.0 21.31.2 
BERT DBERTtest 0.0 5.3 18.91.2 29.41.1 
RoBERTa DRoBERTatest 0.0 5.9 16.10.8 26.70.9 
OriginalRe-init.
ModelResourceEMF1EMF1
BiDAF DBiDAFdev 0.0 5.3 10.70.8 20.41.0 
BERT DBERTdev 0.0 4.9 19.71.0 30.11.2 
RoBERTa DRoBERTadev 0.0 6.1 15.70.9 25.81.2 
 
BiDAF DBiDAFtest 0.0 5.5 11.61.0 21.31.2 
BERT DBERTtest 0.0 5.3 18.91.2 29.41.1 
RoBERTa DRoBERTatest 0.0 5.9 16.10.8 26.70.9 
Close Modal

or Create an Account

Close Modal
Close Modal