avatar

Hyunjin Kim

Ph.D. Student
KAIST School of Computing
hyunjin@kaist.ac.kr


About Me

I am Hyunjin Kim, a PhD student at KAIST School of Computing, advised by Prof. Jae-Gil Lee. I received both my Masters and Bachelors degree at KAIST.

My research focuses on efficient and reliable large language models, with an emphasis on data-centric learning and inference optimization. I study how to reduce computational overhead during inference while maintaining strong performance, particularly in long-context settings.

News

Publications

  1. EMNLP
    Clustering-Based Balanced Sampling and Allocation with Data Parallelism for High-Performance Fine-Tuning
    Hyunjin Kim, Youngeun Nam, Jaemin Han, Wonhyeok Choi, Jae-Gil Lee
    Conference on empirical methods in natural language processing (EMNLP), 2026.
    PDF Code

  2. ArXiv
    TimelyFreeze: Adaptive Parameter Freezing Mechanism for Pipeline Parallelism
    Seonghye Cho, Jaemin Han, Hyunjin Kim, Euisoo Jung, Jae-Gil Lee
    ArXiv, 2026.
    PDF

  3. CVPR
    Accelerating Diffusion via Hybrid Data-Pipeline Parallelism Based on Conditional Guidance Scheduling
    Euisoo Jung, Byunghyun Kim, Hyunjin Kim, Seonghye Cho, Jae-Gil Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
    PDF Code

Services

Teaching


Powered by Jekyll and Minimal Light theme.