BLAST: Block-Level Adaptive Structured Matrices for Efficient Deep Neural Network Inference

Devs

BLAST: Block-Level Adaptive Structured Matrices for Efficient Deep Neural Network Inference | Read Paper on Bytez