Open access
2026
AWa: An Activation-Weight Distribution-Aware Binarization Method for Large Language Models
AWa (Activation-Weight distribution-aware), a novel binarization framework that analyzes the distributions of activations and weights to perform binarization in a systematic, stage-wise manner, is proposed and achieves superior zero-shot average accuracy.
Aeri Kim, Sang-Seol Lee, Jinwoo Jeong et al.
· IEEE Access · 0 citations