Chanmi Lee

Ph.D. Candidate at KAIST

I am a Ph.D. student in the School of Computing at KAIST, affiliated with the SGVR Lab under the supervision of Prof. Sung-Eui Yoon.

Guided by a vision of seamless human-robot symbiosis, my research focuses on developing trustworthy Embodied AI and Robot Foundation Models (RFMs). Specifically, I am dedicated to enhancing model robustness against adversarial attacks and ensuring safe operation in Out-of-Distribution (OOD) environments.

I am always open to academic exchange and collaboration, feel free to reach out!

Chanmi Lee

Education

News

Selected Publications

ICRA26-3DAdvObj Paper

Beyond the Patch: Exploring Vulnerabilities of Visuomotor Policies via Viewpoint-Consistent 3D Adversarial Object

Chanmi Lee, Minsung Yoon, Woojae Kim, Sebin Lee, and Sung-Eui Yoon

IEEE International Conference on Robotics and Automation (ICRA), 2026

RiTA25-Frontier Paper

Mode-Switching Frontier Search for Robust Autonomous Exploration in Complex Terrain

Jinhwan Seo, Chanmi Lee, Hyunsik Son, and Sung-Eui Yoon

International Conference on Robot Intelligence Technology and Applications (RiTA), 2025

ICCV25-Corruption Paper

Towards Robustness of Person Search against Corruptions

Woojung Son, Yoonki Cho, Guoyuan An, Chanmi Lee, and Sung-Eui Yoon

IEEE/CVF International Conference on Computer Vision (ICCV), 2025

IV24-OffRoad Paper

Analysis of Terrain-Aware Optimal Path Planning Methods for Stable Off-Road Navigation

Minsung Yoon, Taegeun Yang, Chanmi Lee, Hyunsik Son, and Sung-Eui Yoon

Off-Road Autonomy Workshop at IEEE Intelligent Vehicles Symposium (IV), 2024

Awards

Archive