An Energy-Efficient High-Utilization Hardware Architecture for Attention Mechanism in Transformer using Balanced Systolic Array and Multi-Row Interleaved Operation Ordering

Haiyang Zhou(Chinese Academy of Sciences), Ming Liu(China Academy of Space Technology), Chunmeng Dou(Chinese Academy of Sciences), Hanghang Gao(Chinese Academy of Sciences), Yuanlu Xie(Institute of Microelectronics), Xiaoxin Xu(Shanghai University), Hongyang Hu(Chinese Academy of Sciences), Jinshan Yue(Chinese Academy of Sciences)
Unknown
June 22, 2025
Cited by 0


Related Papers