Pruning Redundant Mappings in Transformer Models via Spectral-Normalized Identity Prior.
Zi Lin, Jeremiah Z. Liu, Zi Yang, Nan Hua, Dan Roth
Browse the full EMNLP paper archive.
Zi Lin, Jeremiah Z. Liu, Zi Yang, Nan Hua, Dan Roth
Browse the full EMNLP paper archive.