1 paper
Cenlin Duan, Jianlei Yang, Yiou Wang +7
Bit-level sparsity in neural network models harbors immense untapped potential. Eliminating redundant calculations of randomly distributed zero-bits significantly boosts computatio…