Zejia Lin(林泽佳)

I am a third-year Ph.D. student of Computer Science at Sun Yat-sen University co-advised by Prof. Yutong Lu and Assoc. Prof. Xianwei Zhang. I completed bachelor degree at Northwestern Polytechnical University.

My research interests lie in optimizing GPU computing systems for general computing and AI workloads. A particular emphasis is on software designs (architecture/compiler/runtime) around resource sharing and mixed-precision computation.

Publications

PDF Slides GoPTX: Fine-grained GPU Kernel Fusion by PTX-level Instruction Flow Weaving
Kan Wu, Zejia Lin, Mengyue Xi, Zhongchun Zheng, Wenxuan Pan, Xianwei Zhang, and Yutong Lu.
ACM/IEEE Design Automation Conference.

PDF Slides MixPert: Optimizing Mixed-Precision Floating-Point Emulation on GPU Integer Tensor Cores
Zejia Lin, Aoyuan Sun, Xianwei Zhang, and Yutong Lu.
ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems.

PDF Slides KeSCo: Compiler-based Kernel Scheduling for Multi-task GPU Applications
Zejia Lin#, Zewei Mo#, Xuanteng Huang, Xianwei Zhang, and Yutong Lu.
IEEE International Conference on Computer Design.

PDF Slides moTuner: A Compiler-based Auto-tuning Approach for Mixed-precision Operators
Zewei Mo, Zejia Lin, Xianwei Zhang, and Yutong Lu.
ACM International Conference on Computing Frontiers.

# Equal contribution.

Experiences

  • [2024.12 – Now]. LLM research intern at Wechat Search, Tencent (Guangzhou).
  • [Spring 2023]. Teaching assistant of DCS290/292: Compilation Principle, Sun Yat-sen University
  • [2021.07 – 2021.09]. Backend develop intern at Wechat Pay, Tencent (Shenzhen).

Awards

  • [2022, 2023]. First Class Scholarship, Sun Yat-sen University.
  • [2022]. Excellence Scholarship, Tencent Technology.
  • [2018 – 2021]. First Class Scholarship, Northwestern Polytechnical University.
  • [2020]. Second Price, CCCC Wechat Mini Program Development.