Table 4: 

F1 scores of BERT-large models fine-tuned on CoNLL and evaluated on randomly permuted versions of the dev and test sets: π(dev) and π(test).

Methodπ(dev)π(test)
BERT-large 23.45 25.46 
 
 +adv 31.98 31.99 
 +adv&mask 35.02 34.09 
 +adv&mask&freeze 40.39 38.62 
Methodπ(dev)π(test)
BERT-large 23.45 25.46 
 
 +adv 31.98 31.99 
 +adv&mask 35.02 34.09 
 +adv&mask&freeze 40.39 38.62 
Close Modal

or Create an Account

Close Modal
Close Modal