Shengwen Liang

State Key Laboratory of Processors (SKLP), Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS)

liangshengwen_pic.jpg

Dr. Shengwen Liang is currently an associate professor in State Key Laboratory of Processors (SKLP), Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS). His research interest includes domain specific storage device and system design, hardware accelerator design for intelligent applications, and disruptive technology for chip design. He has published over 20 papers at leading conferences and journals, including ISCA, MICRO, HPCA, ATC, DAC, and IEEE TC. Among his awards are the Best Paper Award at the IEEE Transactions on Computers (2021) and the IEEE Asian Test Symposium (2023), the Huawei Olympus Pioneer Award (2024), and the third prize of FPGA track in Low-Power Computer Vision Challenge (2020).

news

Jan 19, 2025 Our work NeuVSA, a unified and efficient architecture for neural vector search, is accepted by HPCA’2025. Congratulations to Ziming Yuan.

Selected publications

  1. APoX-M: Accelerate deep point cloud analysis via adaptive graph construction
    Lei Dai, Shengwen Liang, Ying Wang, Huawei Li, and Xiaowei Li
    Integration, the VLSI Journal, 2025
  2. AGC: A Unified Architecture for Accelerating K-Nearest Neighbor Graph Construction in Vector Search
    Lei Dai, Ziming Yuan, Wen Li, Shengwen Liang, Kaiwei Zou, Ying Wang, Cheng Liu, Huawei Li, and Xiaowei Li
    In 2024 IEEE/ACM International Conference on Computer Aided Design (ICCAD), New Jersey, USA, Oct 2024
  3. Flagger: Cooperative acceleration for large-scale cross-silo federated learning aggregation
    Xiurui Pan, Yuda An, Shengwen Liang, Bo Mao, Mingzhe Zhang, Qiao Li, Myoungsoo Jung, and Jie Zhang
    In 2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA), Buenos Aires, Argentina, Oct 2024
  4. Cambricon-LLM: A Chiplet-Based Hybrid Architecture for On-Device Inference of 70B LLM
    Zhongkai Yu, Shengwen Liang, Tianyun Ma, Yunke Cai, Ziyuan Nan, Di Huang, Xinkai Song, Yifan Hao, Jie Zhang, Tian Zhi, Yongwei Zhao, Zidong Du, Xing Hu, Qi Guo, and Tianshi Chen
    In 2024 57th IEEE/ACM International Symposium on Microarchitecture (MICRO), Austin, TX, USA, Oct 2024
  5. HyQA: Hybrid Near-Data Processing Platform for Embedding Based Question Answering System
    Shengwen Liang, Ziming Yuan, Ying Wang, Dawen Xu, Huawei Li, and Xiaowei Li
    In 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE), Valencia, Spain, Mar 2024