BERT and PALs Projected Attention Layers for Efficient Adaptation in Multi-Task Learning

BERTandPALs:ProjectedAttentionLayersforEfficientAdaptationinMulti-TaskLearningAsaCooperStickland1IainMurray1AbstractHowever,fine-tuningseparatemodelsforeachtaskoftenworksbetterinpractice.Althoughwe...