Sunung Mun

Machine Learning Lab · POSTECH

Sunung Mun

Ph.D. candidate at the Graduate School of AI, POSTECH, researching controllable multimodal generation and editing for adaptive visual content.

Email Google Scholar LinkedIn Resume

Research

Controllable generative models for adaptive visual content.

My research focuses on making pretrained multimodal generative models more controllable, reliable, and useful in real-world settings. I study diffusion-based image and video editing, compositional generation, and agentic refinement methods that diagnose and correct generation failures while preserving important visual details.

Selected Work

Publications

ICCV 2025

Addressing Text Embedding Leakage in Diffusion-based Image Editing

Sunung Mun*, Jinhwan Nam*, Sunghyun Cho, Jungseul Ok

International Conference on Computer Vision, 2025

* Equal contribution

CVPR 2024

MedBN: Robust Test-Time Adaptation against Malicious Test Samples

Hyejin Park*, Jeongyeon Hwang*, Sunung Mun, Sangdon Park, Jungseul Ok

IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

* Equal contribution

WHC 2019

Perceptual Space of Regular Homogeneous Haptic Textures Rendered Using Electrovibration

Sunung Mun, Hojin Lee, Seungmoon Choi

IEEE World Haptics Conference, 2019

CHI 2018

Experiences Enabled by Automatic Synthesis of Motion and Vibrotactile Effects

Sangyoon Han, Sunung Mun, Jongman Seo, Jaebong Lee, Seungmoon Choi

Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, 2018

CHI 2018

Substituting Motion Effects with Vibrotactile Effects for 4D Experiences

Jongman Seo, Sunung Mun, Jaebong Lee, Seungmoon Choi

Proceedings of the CHI Conference on Human Factors in Computing Systems, 2018

Manuscripts

Preprints

arXiv 2026

EPIC: Efficient Predicate-Guided Inference-Time Control for Compositional Text-to-Image Generation

Sunung Mun, Sunghyun Cho, Jungseul Ok

Under review

arXiv 2026

Towards Faithful Agentic XAI: A Verification Method and an Open-World Benchmark for Better Model Faithfulness

Jaechang Kim, Sunung Mun, Seungjoon Lee, Jaewoong Cho, Jungseul Ok

Under review

Academic Service

Reviewer

2026

Conference Reviewing

Reviewer for CVPR 2026, ECCV 2026, and NeurIPS 2026.

Background

Education

Sep 2022 - Present

Ph.D. Candidate, Artificial Intelligence

POSTECH, Machine Learning Lab

Supervised by Prof. Jungseul Ok · GPA: 4.17/4.3

2017 - 2019

M.S., Computer Science

POSTECH, Haptics and Virtual Reality Lab

Supervised by Prof. Seung-Moon Choi · GPA: 3.78/4.3

2012 - 2017

B.S. in Computer Science and Engineering

POSTECH

Magna Cum Laude · GPA: 3.7/4.3

Industry

Work Experience

May 2021 - Aug 2022

Lunit · Frontend Engineer

  • Developed a frontend for a data annotation tool used to collect data for machine learning training.
  • Built a frontend for a project management system.

Apr 2019 - May 2021

Tmax · Frontend Engineer / VR Team Lead

  • Developed the cloud storage module frontend for Tmax Waffle.
  • Built a web-based video conferencing solution.
  • Led the Tmax OS VR Team, conducting VR/AR market research and rendering-technology studies.

Toolkit

Skills