A 22-nm 64-kB lightning-like hybrid computing-in-memory macro with a compressed adder tree and analog-storage quantizers for transformer and CNNs

An Guo 0001, Xi Chen 0107, Fangyuan Dong, Jinwu Chen, Zhihang Yuan, Xing Hu 0010, Guangyu Sun 0003, Xiaomin Li, Arindam Basu, Jun Yang 0006, Xin Si. A 22-nm 64-kB lightning-like hybrid computing-in-memory macro with a compressed adder tree and analog-storage quantizers for transformer and CNNs. Science in China Series F: Information Sciences, 68(12), 2025. [doi]

Abstract

Abstract is missing.