Skip to Main Content
Table 13:ȃ
Cross-model micro-averaged accuracy (%). The first row is the model to test, and the second row is the model on which prompt weights are learned.
TestBERTERNIE
TrainBERTERNIEERNIEBERT
Mine 38.9 38.0 42.3 38.7 
Mine+Man 39.6 39.5 43.8 40.5 
Mine+Para 36.2 34.2 40.1 39.0 
Man+Para 37.3 35.2 41.1 40.3 
TestBERTERNIE
TrainBERTERNIEERNIEBERT
Mine 38.9 38.0 42.3 38.7 
Mine+Man 39.6 39.5 43.8 40.5 
Mine+Para 36.2 34.2 40.1 39.0 
Man+Para 37.3 35.2 41.1 40.3 
Close Modal

or Create an Account

Close Modal
Close Modal