Junwha Hong

I am a Research Officer at the Agency for Defense Development (ADD), working on LLM serving in heterogeneous clusters and efficient vision inference.

My current research is advised by Prof. Olivia Hsu (CMU) and Prof. Jongse Park (KAIST). Previously, I worked on memory bug detection and fuzzing at UNIST S2Lab under Prof. Yuseok Jeon (Korea Univ.).

Publications

IEEE S&P 2025 acceptance rate: 14.8%

Junwha Hong, Wonil Jang, Mijung Kim, Lei Yu, Yonghwi Kwon, Yuseok Jeon

The first CMA-aware address sanitizer designed to detect memory bugs in objects allocated by custom memory allocators. CMASan identified 19 security vulnerabilities, including 7 CVEs: CVE-2023-7104, CVE-2023-7152, CVE-2023-7158, CVE-2024-8946, CVE-2024-8947, CVE-2024-8948, and CVE-2024-11235.

Paper Code

NeurIPS 2025 acceptance rate: 24.52%

Yunghee Lee, Byeonghyun Pak, Junwha Hong, Hoseong Kim

A training-free strategy for accelerating diffusion sampling by treating classifier-free guidance as a multirate system.

Project Paper Code

Mar 2024 - Present

Research Officer for National Defense (First Lieutenant)

Manager: Dr. Eunjin Koh

Dec 2020 - Nov 2024

Undergraduate Researcher

Advisor: Prof. Yuseok Jeon (Korea Univ.)

Apr 2023 - May 2023

Visiting Student Researcher

Research visit with Prof. Yonghwi Kwon (now UMD)

Mar 2020 - Feb 2024

B.S. in Computer Science and Engineering

GPA: 4.13/4.3, summa cum laude (ranked 4th out of 296 students).

Activities: 11th President of HeXA, Hacker's eXciting Academy.