Skip to Main Content
Table 7: 
Accuracy of beginning ranking and logic ranking. Larger scores are better.
ModelsBR (%)LR (%)
GPT-2 (Pretrain) 59.3 44.8 
GPT-2 (Fine-tune) 73.4 69.6 
 
Ours 76.2 72.7 
w/o Knowledge 74.9 71.5 
w/o Multi-task 75.7 70.4 
ModelsBR (%)LR (%)
GPT-2 (Pretrain) 59.3 44.8 
GPT-2 (Fine-tune) 73.4 69.6 
 
Ours 76.2 72.7 
w/o Knowledge 74.9 71.5 
w/o Multi-task 75.7 70.4 
Close Modal

or Create an Account

Close Modal
Close Modal