"Guanghui" in Chinese means "sheen" in English

I am Guanghui Song, a PhD student at Hunan University under the supervision of Prof. Jie Zhao. I completed my master's degree at Information Engineering University and then worked as an AI Compiler R&D Engineer at Li Auto Inc., before starting my PhD.

• Advanced Compilation Techniques: Polyhedral analysis and scheduling, affine loop optimizations, and dataflow-oriented mapping from loop nests and graphs to efficient forms — often through LLVM/MLIR-based passes and reusable compiler pipelines.
• Hardware Acceleration: Hardware-aware compilation and mapping for NPUs and spatial and reconfigurable architectures (CGRA, FPGA), and HLS-oriented flows—covering operator lowering, pipelining, and resource utilization from IR down to accelerator execution.
• High Performance Computing: Mixed-precision code generation and autotuning for numerical kernels, lightweight analytical performance models, and scalable execution strategies on parallel and specialized hardware.

My research has appeared in venues such as ISCA, PPoPP, TACO, ASE, and the Journal of Software.

I'm always interested in collaboration and discussion about my research areas. Feel free to reach out:

• 📧 Email: sheensong@hnu.edu.cn
• 🌐 Academic Website: sheensong.top/academic

• Mar 2026: Paper "Dynamic Scheduling for AI Accelerators via TISA" accepted by ISCA 2026
• Mar 2026: Paper "A Decoupled Analytical Model for Tile Size Selection in Affine Programs" accepted by ACM TACO
• Sep 2025: Began a research internship at the National University of Singapore (NUS) School of Computing as a visiting research intern
• Jun 2025: Attended PLDI 2025 conference in South Korea, supported by the PLMW @ PLDI 2025 scholarship
• Apr 2025: Moderated Session 6 (Student Talks and Technical Talk) at EuroLLVM 2025 in Berlin, supported by a student travel grant from the LLVM Foundation
• Dec 2024: Ran for president of the CCF Hunan University Student Chapter
• Sep 2024: Started PhD studies at Hunan University after leaving Li Auto Inc.
• Jul 2024: Presented research at CCF Chip 2024 conference
• Mar 2024: Presented paper at PPoPP 2024 conference in Edinburgh
• Nov 2023: Paper "A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine Programs" accepted by PPoPP 2024
• Jul 2023: Graduated with Master's degree from Information Engineering University and joined Li Auto Inc. as AI Compiler R&D Engineer
• Aug 2022: Paper "Automatic Mixed Precision Optimization for Stencil Computation" accepted in Journal of Software (Master's thesis work)