Scholar
Dongjae Jeon
Google Scholar ID: SaaN_bAAAAAJ
Yonsei University
Citations & Impact
Citations
28
H-index
2
i10-index
2
Publications
9
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - 'Rainbow Padding: Mitigating Early Termination In Instruction-tuned Diffusion LLMs' (Preprint)
- - 'A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models' (Preprint)
- - 'Preserve then Quantize: Dominant-Subspace Guided Low-Rank Reconstruction' (ICML TTODLer-FM workshop, Oral)
- - 'Information-Theoretic Discrete Diffusion' (NeurIPS)
- - 'Assigning Distinct Roles to Quantized and Low-Rank Matrices Toward Optimal Weight Decomposition' (ACL Findings)
- - 'Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes' (ICML, Spotlight)
- - 'Large Language Models Still Exhibit Bias in Long Text' (ACL Findings)
- - 'An Information Theoretic Metric for Evaluating Unlearning Models' (Preprint)
- Awards:
- - 2nd place in CVPR CLVISION challenge
- - 1st place in ICCV VCL challenge
Research Experience
- Conducted research on Continual Learning in the context of computer vision; Gave a talk at Yonsei MLSys student group on Diffusion models & their privacy issues.
Education
- Master's: Computer Science at Yonsei University, Advisor: Albert No
Background
- First-year master's student in Computer Science at Yonsei University, focusing on safe and reliable AI, particularly generative models. Currently interested in advancing Diffusion Language Models (sampling methodologies and architectures) and developing efficient AI systems.
Miscellany
- Personal interests not specified
