Seonghwan Park

I am a researcher at the Korea Electronics Technology Institute (KETI). I received my M.S. from POSTECH under the supervision of Prof. Namhoon Lee.

My research interests focus on improving model efficiency and robustness from an optimization perspective. I am interested in multi-modal foundation models, as well as research in 3D vision and physical AI.

Email / Google Scholar / LinkedIn

Publications

(Tentative Title) Efficient Out-of-Distribution Detection Method
Seonghwan Park, Hyunji Jung, Dongyeop Lee, Namhoon Lee
Under review

An Analysis of Concept Bottleneck Models: Measuring, Understanding, and Mitigating the Impact of Noisy Annotations
Seonghwan Park, Jueun Mun, Donghyun Oh, Namhoon Lee
NeurIPS, 2025

OAIG: Object-Centric Visual Prompting for Vision-Language Models
Jaehyeon Jeong, Seonghwan Park, Namhoon Lee
Conference of Korean Artificial Intelligence Association (CKAIA), 2025
Best Paper Finalist Award

ZIP: An Efficient Zeroth-order Prompt Tuning for Black-box Vision-Language Models
Seonghwan Park, Jaehyeon Jeong, Yongjun Kim, Jaeho Lee, Namhoon Lee
ICLR, 2025

FedFwd: Federated Learning without Backpropagation
Seonghwan Park, Dahun Shin, Jinseok Chung, Namhoon Lee
ICML Workshop on Federated Learning and Analytics in Practice: Algorithms, Systems, Applications, and Opportunities, 2023

On the Effectiveness of Sharpness-Aware Minimization with Large Mini-batches
Jinseok Chung, Seonghwan Park, Jaeho Lee, Namhoon Lee
ICML Workshop on High-dimensional Learning Dynamics, 2023

Education

Pohang University of Science and Technology (POSTECH) Master of Science in Artificial Intelligence, advised by Professor Namhoon Lee Total GPA of 3.9 / 4.0 (4.0 / 4.3)	Mar. 2023 – Aug. 2025 Pohang, Korea
Hanyang University Bachelor of Science in Computer Science and Engineering Total GPA of 3.86 / 4.0 (4.12 / 4.5)	Mar. 2017 – Feb. 2023 Seoul, Korea

Work Experience

Researcher Korea Electronics Technology Institute (KETI) Research Area: 3D Vision, Physical AI	Jan. 2026 - Now
Research Assistant Computational Optimization Lab, POSTECH Research Area: Optimization, Multi-modal Foundation Models Supervisor: Prof. Namhoon Lee	Aug. 2025 - Dec. 2025
Samsung Research Project Research Area: Multi-modal Foundation Models for Anomaly Detection	Jun. 2025 - Nov. 2025
Samsung Advanced Institute of Technology (SAIT) Project Research Area: Multi-modal Foundation Models, Visual Prompt Tuning	Jun. 2024 - May. 2025
Undergraduate Research Intern Machine Learning Lab, Hanyang University Research Area: Large-scale Language Models Supervisor: Prof. Eun-Sol Kim	Jan. 2022 – Dec. 2022

Teaching Experience

POSTECH Teaching Assistant Topic: Introduction to Artificial Intelligence	Sep. 2024 – Dec. 2024
POSCO AI Fellowship Teaching Assistant Topic: Deep Learning	May 2023 – Jun. 2023

Honors and Awards

CKAIA Best Paper Finalist Award	2025
HYU Academic Excellence Award (within top 3%)	2022

Service

Reviewers ICLR TMLR	2026
Republic of Korea Air Force (ROKAF), Logistics Command Served as Sergeant	Oct. 2018 – Aug. 2020

webpage template from Jon Barron