Dongjae Jeon


I am a first-year master's student in Computer Science at Yonsei University, working in the Artificial Intelligence and Information Systems Lab with Professor Albert No.

My research focuses on safe and reliable AI, with an emphasis on generative models. Currently, I am interested in advancing Diffusion Language Models (sampling methodologies and architectures). I also pursue the development of efficient AI systems. Previously, I have conducted research on Continual Learning in the context of computer vision. My primary goal is to deepen our understanding of how machines perceive and to create methods that enhance their reliability.

Here is my CV. Feel free to contact me via email if you are interested.

Bio Mail Scholar Github LinkedIn

Profile picture

News


Sep 2025: A paper on Diffusion Language Model is accepted to NeurIPS 2025. Kudos to my co-authors! See you in San Diego πŸ‡ΊπŸ‡Έ!
Aug 2025: Official code for ICML 2025 paper is uploaded! Checkout!
Aug 2025: I gave a talk at Yonsei MLSys student group on Diffusion models & their privacy issues.
Jun 2025: A paper on Quantization is accepted to ICML 2025 (TTODLer-FM workshop) as Oral!
May 2025: Two Papers (ODLRI and Regard-FT) are accepted to ACL 2025! See you in Austria πŸ‡¦πŸ‡Ή!
May 2025: A paper on Diffusion Safety is accepted to ICML 2025 as Spotlight (top 2.6% = 313/12,107)! See you in Canada πŸ‡¨πŸ‡¦!

Publications


Project image
A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models Safety DGMs
Wonje Jeung*, Sangyeon Yoon*, Yoonjun Cho, Dongjae Jeon, Sangwoo Shin, Hyesoo Hong, Albert No
Preprint, 2025
[arXiv]
Project image
Preserve then Quantize: Dominant-Subspace Guided Low-Rank Reconstruction Efficient ML
Yoonjun Cho*, Dongjae Jeon*, Soeun Kim, Albert No
ICML (TTODLer-FM workshop), 2025 (Oral)
[arXiv] [slide]
Project image
Information-Theoretic Discrete Diffusion DGMs
Moongyu Jeon, Sangwoo Shin, Dongjae Jeon, Albert No
NeurIPS, 2025
[arXiv]
Project image
Assigning Distinct Roles to Quantized and Low-Rank Matrices Toward Optimal Weight Decomposition Efficient ML
Yoonjun Cho, Soeun Kim, Dongjae Jeon, Kyelim Lee, Beomsoo Lee, Albert No
ACL Findings, 2025
[arXiv]
Project image
Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes Safety DGMs
Dongjae Jeon*, Dueun Kim*, Albert No
ICML, 2025 (Spotlight)
prelim @ AAAI (PPAI workshop), 2025 (Oral)
[arXiv] [slide] [code]
Project image
Large Language Models Still Exhibit Bias in Long Text Safety
Wonje Jeung, Dongjae Jeon, Ashkan Yousefpour, Jonghyun Choi
ACL Findings, 2025
prelim @ NeurIPS (SOLAR workshop), 2024
[arXiv]

Awards


Project image
Class-Incremental with Repetition (CIR) using Unlabelled Data
2nd place πŸ”₯
CVPR CLVISION challenge, 2024
Project image
Continual Test-time Adaptation for Object Detection
1st place πŸ”₯
ICCV VCL challenge, 2023
[slide] / [report]

Talks


Project image
Privacy Issues in DGMs: How to detect & mitigate
MLSys Student Group Seminar, 2025
[slide]

Template borrowed from Here