Donghyun Lee

Ph.D. Student, Computer Science and Engineering, Seoul National University

Email
Curriculum Vitae

Hello! I am a 4th year Ph.D student in the Department of Computer Science and Engineering at Seoul National University. I am a member of Architecture and Code Optimization (ARC) Lab led by Professor Jae W. Lee.

My research focuses on systems for machine learning, with current interests in efficient 3D vision and GPU kernel optimization.

I received my B.S. degree in Electrical and Electronic Engineering from Yonsei University in 2022.

News

Jan 17, 2026 Paper accepted to ASPLOS 2026. GS-Scale: Unlocking Large-Scale 3D Gaussian Splatting Training via Host Offloading
June 26, 2025 Paper accepted to ICCV 2025. FastPoint: Accelerating 3D Point Cloud Model Inference via Sample Point Distance Prediction
July 1, 2024 Paper accepted to ECCV 2024. Frugal 3D Point Cloud Model Training via Progressive Near Point Filtering and Fused Aggregation

Publications

  1. ASPLOS
    GS-Scale: Unlocking Large-Scale 3D Gaussian Splatting Training via Host Offloading
    Donghyun Lee, Dawoon Jeong, Jae W. Lee, and Hongil Yoon
    31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2026
    PDF Code
  2. ICCV
    FastPoint: Accelerating 3D Point Cloud Model Inference via Sample Point Distance Prediction
    Donghyun Lee, Dawoon Jeong, Jae W. Lee, and Hongil Yoon
    IEEE/CVF International Conference on Computer Vision (ICCV) 2025
    PDF Code Slides Talk
  3. HPCA
    FACIL: Flexible DRAM Address Mapping for SoC-PIM Cooperative On-device LLM Inference
    Seong Hoon Seo, Junghoon Kim, Donghyun Lee, Seonah Yoo, Seokwon Moon, Yeonhong Park, and Jae W. Lee
    31st IEEE International Symposium on High-Performance Computer Architecture (HPCA) 2025
    PDF
  4. ECCV
    Frugal 3D Point Cloud Model Training via Progressive Near Point Filtering and Fused Aggregation
    Donghyun Lee, Yejin Lee, Jae W. Lee, and Hongil Yoon
    European Conference on Computer Vision (ECCV) 2024
    PDF Code Slides Talk
  5. APSys
    Liquid: Mix-and-Match Multiple Image Formats to Balance DNN Train- ing Pipeline
    Woohyeon Baek, Jonghyun Bae, Donghyun Lee, Hyunwoong Bae, Yeonhong Park, and Jae W. Lee
    14th ACM SIGOPS Asia-Pacific Workshop on Systems (APSys) 2023
    PDF
  6. AAAI
    Not All Neighbors Matter: Point Distribution-Aware Pruning for 3D Point Cloud
    Yejin Lee, Donghyun Lee, JungUk Hong, Jae W. Lee, and Hongil Yoon
    37th AAAI Conference on Artificial Intelligence (AAAI) 2023
    PDF Code Slides

Honors and Awards

2025 The 31st Samsung Humantech Paper Award: Encouragement Prize - Samsung Electronics Co., Ltd.
2024 Accelerator Programming Winter School - CUDA competition, 1st place, SNU THUNDER Research Group & Manycoresoft
2022 Highest Honors at Graduation (Top 1% of class) - Yonsei University

Open-Source

[ICCV'25] FastPoint: Accelerating 3D Point Cloud Model Inference via Sample Point Distance Prediction
https://github.com/SNU-ARC/FastPoint
[ECCV'24] Frugal 3D Point Cloud Model Training via Progressive Near Point Filtering and Fused Aggregation
https://github.com/SNU-ARC/Frugal_PN_Training

Services

2022 Student Volunteer
International Symposium on Code Generation and Optimization (CGO)

Teaching Experience

Sep-Dec, 2023 Teaching assistant for Computer Architecture Course (Instructor: Prof. Jae W. Lee)
Course Summary: Learning computer architecture including ISA, CPU, Pipeline, Memory hierarchy.