Enhancing Ultra-Low-Bit Quantization of Large Language Models Through Saliency-Aware Partial Retraining.

Deyu Cao, Samin Aref

Browse the full MDAI paper archive.