uanu2002/JSQ

[ICML 2024] JSQ: Compressing Large Language Models by Joint Sparsification and Quantization