Jinhyeok Kim | Seoul National University

About Me

I am an integrated M.S./Ph.D. student in Electrical and Computer Engineering at Seoul National University, advised by Prof. Jaeyoung Do in the AIDAS Lab. Previously, I received my B.S. in Electrical and Electronic Engineering from Yonsei University.

My research focuses on efficient AI systems, particularly LLM inference and serving, GPU kernel optimization, and algorithm–system co-design.

Research Interests

Efficient inference: LLM inference and serving, sparsity, and model compression
GPU systems: Kernel optimization and algorithm–system co-design

News

[Sep. 2026] Our paper, Dynin-Robotics, was accepted to NeurIPS 2026!
[Jul. 2026] Our paper, Unified Static-Dynamic Pruning for Efficient LLM Inference (SPDP), was accepted to VLDB 2026!

Publications

NeurIPS

Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language Action Model

Hoeun Lee, Jaeik Kim, Jusang Oh, Jinhyeok Kim, Geon Choi, Hyeonggeun Kim, Jaeyoung Do^†

40th Annual Conference on Neural Information Processing Systems (NeurIPS), 2026

Paper Project Model Code
VLDB

Unified Static-Dynamic Pruning for Efficient LLM Inference

Jinhyeok Kim, Yejoon Lee, Jaeyoung Do^†

Proceedings of the VLDB Endowment (VLDB), 2026

Paper Code
arXiv

Dynin-Omni: Omnimodal Unified Large Diffusion Language Model

Jaeik Kim, Woojin Kim, Jihwan Hong, Yejoon Lee, Sieun Hyeon, Mintaek Lim, Yunseok Han, Dogeun Kim, Hoeun Lee, Hyunggeun Kim, Jinhyeok Kim, Jaeyoung Do^†

arXiv preprint, 2026

Paper Project Model Code

^† Corresponding author

Education

Seoul National University · Mar. 2025 – Present
Integrated M.S./Ph.D. in Electrical and Computer Engineering
Advisor: Prof. Jaeyoung Do
Yonsei University · Mar. 2019 – Feb. 2025
B.S. in Electrical and Electronic Engineering

Research Experience

Graduate Research Assistant, AIDAS Lab, Seoul National University
Mar. 2025 – Present
Research Intern, AIDAS Lab, Seoul National University
Jul. 2024 – Feb. 2025
Research Intern, Information Communications & Security Lab, Yonsei University
Jan. 2024 – Feb. 2024

Technical Skills

Programming & ML: CUDA, C/C++, Python, PyTorch, Bash
Hardware & tools: Verilog, RISC-V, MATLAB, Xilinx Vivado, ModelSim, Design Compiler