publications

2026

  1. ISCA
    SLICE: A Selective Local Inference Framework with Codec Exploitation for Accelerating Video Super-Resolution
    Mingu Jung, Sungbin Kim, Seunghyun Lee, and 3 more authors
    In The 53rd Annual International Symposium on Computer Architecture (ISCA), 2026
  2. ISCA
    MXFFP: Microscaling Flexible Floating Point Format for Large-Scale AI Model Acceleration
    Sungwoo Kim, Sungbin Kim, Dongho Ha, and 6 more authors
    In The 53rd Annual International Symposium on Computer Architecture (ISCA), 2026
  3. Journal
    DeSpa: Heterogeneous multi-core accelerators for energy-efficient dense and sparse computation at the tile level in Deep Neural Networks
    Hyungjun Jang, Dongho Ha, Hyunwuk Lee, and 1 more author
    Journal of Systems Architecture, 2026
    LINK

2025

  1. US Patent
    Apparatus and method with register sharing
    Won Woo Ro, Seunghyun Jin, Jonghyun Lee, and 1 more author
    2025
    US Patent 12,298,901
    LINK
  2. MICRO
    BitL: Hybrid Bit-Serial and Parallel Processing for Critical Path Reduction
    Seunghyun Lee, Dongho Ha, Sungbin Kim, and 3 more authors
    In 58th IEEE/ACM International Symposium on Microarchitecture (MICRO 2025), 2025
    LINK
  3. ICCV
    Adversarial Purification via Super-Resolution and Diffusion
    Mincheol Park, Cheonjun Park, Seungseop Lim, and 4 more authors
    In International Conference on Computer Vision (ICCV 2025), 2025
    LINK
  4. ISPASS
    COSMOS: An LLC Contention Slowdown Model for Heterogeneous Multi-core Systems
    Yongju Lee, Jaewon Kwon, Cheolhwan Kim, and 4 more authors
    In 2025 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2025
    LINK
  5. DAC
    CVMAX: Accelerator Architecture with Polar Form Multiplication for Complex-Valued Neural Networks
    Hyunwuk Lee, Sungbin Kim, Sungwoo Kim, and 1 more author
    In The 62nd Design Automation Conference (DAC), 2025
    LINK
  6. HPCA
    Ditto: Accelerating Diffusion Model via Temporal Value Similarity
    Sungbin Kim*, Hyunwuk Lee*, Wonho Cho, and 2 more authors
    In 2025 IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2025
    arXiv LINK
  7. Journal
    REC: Enhancing fine-grained cache coherence protocol in multi-GPU systems
    Gun Ko, Jiwon Lee, Hongju Kal, and 2 more authors
    Journal of Systems Architecture, 2025
    LINK

2024

  1. ICCD
    AirGun: Adaptive Granularity Quantization for Accelerating Large Language Models
    Sungbin Kim, Hyunwuk Lee, Sungwoo Kim, and 2 more authors
    In International Conference on Computer Design (ICCD), 2024
    LINK
  2. ISLPED
    GUMSO: Gating Unnecessary On-Chip Memory Slices for Power Optimization on GPUs
    Seunghyun Jin, Hyunwuk Lee, Junsung Kim, and 1 more author
    In International Symposium on Low Power Electronics and Design (ISLPED), 2024
    LINK
  3. Journal
    SHREG: Mitigating register redundancy in GPUs
    Seunghyun Jin, Hyunwuk Lee, Jonghyun Lee, and 2 more authors
    Journal of Systems Architecture, 2024
    LINK

2023

  1. MICRO
    Exploiting Inherent Properties of Complex Numbers for Accelerating Complex Valued Neural Networks
    Hyunwuk Lee, Hyungjun Jang, Sungbin Kim, and 3 more authors
    In 2023 56th IEEE/ACM International Symposium on Microarchitecture (MICRO), 2023
    LINK
  2. ISPASS
    Early-Adaptor: An Adaptive Framework forProactive UVM Memory Management
    Seokjin Go, Hyunwuk Lee, Junsung Kim, and 3 more authors
    In 2023 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2023
    LINK

2022

  1. US Patent
    Memory device including a plurality of area having different refresh periods, memory controller controlling the same and memory system including the same
    Won Woo Ro, Hyunwuk Lee, Gun Ko, and 4 more authors
    Mar 2022
    US Patent 11,276,452
    LINK

2021

  1. Book
    Deep learning with GPUs
    Won Jeon, Gun Ko, Jiwon Lee, and 3 more authors
    In Advances in Computers, Mar 2021
    LINK