Hongsin Lee

Ph.D. Student
KAIST
hongsin04@kaist.ac.kr

News

[May. 2026] One paper is accepted to ICML 2026.
[May. 2026] One paper is accepted to TMLR.
[Jan. 2025] Two papers are accepted to ICLR 2025.
[Dec. 2024] One paper is accepted to AAAI 2025.

Publications

Toward Understanding Adversarial Distillation: Why Robust Teachers Fail

Hongsin Lee, Hye Won Chung

International Conference on Machine Learning (ICML), 2026

PDF OpenReview Code(Github)
Sample-wise Adaptive Weighting for Transfer Consistency in Adversarial Distillation

Hongsin Lee, Hye Won Chung

Transactions on Machine Learning Research (TMLR), 2026

PDF OpenReview Code(Github)
Indirect Gradient Matching for Adversarial Robust Distillation

Hongsin Lee*, Seungju Cho*, Changick Kim (*Equal contribution)

International Conference on Learning Representations (ICLR), 2025.

PDF OpenReview Code(Zipfile)
Long-tailed Adversarial Training with Self-Distillation

Seungju Cho*, Hongsin Lee*, Changick Kim (*Equal contribution)

International Conference on Learning Representations (ICLR), 2025.

PDF OpenReview Code(Zipfile)
Enhancing Robustness in Incremental Learning with Adversarial Training

Seungju Cho*, Hongsin Lee*, Changick Kim (*Equal contribution)

AAAI Conference on Artificial Intelligence (AAAI), 2025

PDF Code(Github)

Education

Ph.D. in Electrical Engineering, KAIST (Mar. 2025 – Present)
M.S. in Electrical Engineering, KAIST (Mar. 2023 – Feb. 2025)
- Thesis: “Indirect Gradient Matching for Adversarial Robust Distillation”
B.S. in Electrical Engineering, KAIST (Mar. 2017 – Feb. 2023)

Powered by Jekyll and Minimal Light theme.