Viet-Anh on Software Logo

What is: Entropy Minimized Ensemble of Adapters?

SourceEfficient Test Time Adapter Ensembling for Low-resource Language Varieties
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

Entropy Minimized Ensemble of Adapters, or EMEA, is a method that optimizes the ensemble weights of the pretrained language adapters for each test sentence by minimizing the entropy of its predictions. The intuition behind the method is that a good adapter weight α\alpha for a test input xx should make the model more confident in its prediction for xx, that is, it should lead to lower model entropy over the input