About

I am an M.S. student at POSTECH, advised by Namhoon Lee. My research focuses on efficient neural networks, with a particular interest in model compression.

I study how to reduce the memory and computational costs of large-scale models while preserving their capabilities. My current work explores pruning and quantization as practical paths toward more efficient models.

News

    Publications

    * Equal contribution

    Hidden in Plain Sight: The Overlooked Significance of Canonical Elements for Extreme LLM Sparsity

    Hyeondo Jang, Kwanhee Lee, Dongyeop Lee, Namhoon Lee

    EMNLP 2026; CKAIA 2026 Best Paper Award

    Advancing Semiconductor Inspection: A New Dataset and Approach for Robust Anomaly Detection with Large Vision Language Models

    Jaehyeon Jeong*, Hyeondo Jang*, Donghyun Oh, Seongeun Kim, Namhoon Lee

    BMVC 2026

    The Unseen Frontier: Pushing the Limits of LLM Sparsity with Surrogate-Free ADMM

    Kwanhee Lee, Hyeondo Jang, Dongyeop Lee, Dan Alistarh, Namhoon Lee

    ICLR 2026; CKAIA 2025 Best Paper Award

    arXivCode

    Industry Projects

    Large Vision Language Models for Anomaly Detection 2026.01 – 2026.07 Samsung AI Center

    Experience

    AI Research Engineer 2024.09 – 2025.01 LG Energy Solution · AI Technology Team, Seoul, South Korea
    Battery System Research Engineer 2023.07 – 2024.08 LG Energy Solution · SOX Algorithm Team, Gwacheon, South Korea
    Student Researcher 2021.09 – 2021.12 KIST (Korea Institute of Science and Technology) · ANSUR Lab

    Education

    POSTECH 2025.03 – Present M.S. student
    Hanyang University 2017.03 – 2023.02 B.S. in Electrical Engineering

    Honors & Awards

    Best Paper Award 2026.07 CKAIA 2026
    Best Paper Award 2025.11 CKAIA 2025
    Kwanjeong Scholarship 2025.04 Kwanjeong Lee Jong-Hwan Educational Foundation
    KEPCO Scholarship 2022.03 Korea Electric Power Corporation
    ECFCSF Scholarship 2021.03 Korea Electric Construction Financial Cooperative Scholarship Foundation