Under Review

  • Scalable Derivative Gaussian Processes via Exact Gradient Reduction
    Hyunseok Seung and Matthias Katzfuss

Conference Proceedings

  • Low-Rank Curvature for Zeroth-Order Optimization in LLM Fine-Tuning
    Hyunseok Seung, Jaewoo Lee, and Hyunsuk Ko
    AAAI Conference on Artificial Intelligence, 2026 Acceptance rate 17.6%
    Paper Code Poster
  • MAC: An Efficient Gradient Preconditioning using Mean Activation Approximated Curvature
    Hyunseok Seung, Jaewoo Lee, and Hyunsuk Ko
    IEEE International Conference on Data Mining (ICDM), 2025 Acceptance rate 22.0%
    Selected as one of the best-ranked papers
    Paper Code
  • NysAct: A Scalable Preconditioned Gradient Descent using Nyström Approximation
    Hyunseok Seung, Jaewoo Lee, and Hyunsuk Ko
    IEEE International Conference on Big Data, 2024
    Paper Code
  • An Adaptive Method Stabilizing Activations for Enhanced Generalization
    Hyunseok Seung, Jaewoo Lee, and Hyunsuk Ko
    IEEE International Conference on Data Mining Workshop (ICDMW), 2024
    Paper Code

Journal Articles

  • Enhancing COVID-19 Mortality Prediction with Online Autocovariance Change Points Detection
    Hyunseok Seung, Kaiwen Han, Ye Shen, and Yuan Ke
    Stat, 2026
  • Mean Activation Curvature for Scalable Second-Order Optimization in Deep Networks
    Hyunseok Seung, Jaewoo Lee, and Hyunsuk Ko
    Knowledge and Information Systems, 2026 IF 3.1
    Invited journal extension of our IEEE ICDM 2025 paper
  • Modified Likelihood Ratio Tests for Extreme Value Distributions
    Hyunseok Seung and Sangun Park
    Communications in Statistics – Theory and Methods, 2023
    Paper