Boost Transformer-based Language Models with GPU-Friendly Sparsity and Quantization.

Chong Yu, Tao Chen, Zhongxue Gan

Browse the full ACL paper archive.