Huixi Technology and Shanghai Jiao Tong University's joint research papers were accepted by DAC 2023

Two technical papers co-authored by Huixi Technology and Shanghai Jiaotong University were accepted by the 2023 Design Automation Conference (DAC), which are titled FLNA: An Energy-Efficient Point Cloud Feature Learning Accelerator with Dataflow Decoupling and COSA: Co-Operative Systolic Arrays for Multi-head Attention Mechanism in Neural Network using Hybrid Data Reuse and Fusion Methodologies.

Efficient transformer inference architecture is the computing foundation for the large-scale commercialization of BEV and GPT, and efficient point cloud data processing architecture helps LIDAR enable safer intelligent driving. These two collaborative efforts by Huixi Technology and SJTU are the crystallization of our technical concept in the field of cutting-edge industrial applications and architecture. We believe that efficient computing power brings inclusive intelligence, and the closed loop of innovation will accelerate this process.