Myunsoo Kim (김면수)
Contact : ✉ m970326@korea.ac.kr Homepages: 🎓Google Scholar, 💼 LinkedIn
I am a 4th-year Ph.D. candidate at Korea University, advised by Prof. Byung-Jun Lee. I received my B.S. in Statistics in February 2023 from Korea University.
Contact : ✉ m970326@korea.ac.kr Homepages: 🎓Google Scholar, 💼 LinkedIn
I am a 4th-year Ph.D. candidate at Korea University, advised by Prof. Byung-Jun Lee. I received my B.S. in Statistics in February 2023 from Korea University.
Research Interests
My primary research interest lies in leveraging Reinforcement Learning to push the boundaries of modern Generative Models, fundamentally improving both model performance and training efficiency.
Language Models
Alignment: [C3]
Retrieval-Augmented Generation: [C4]
Agentic AI: [W1, P1]
Multimodal Models
Alignment: [C5]
Diffusion Models
Efficient Training: [C1]
Reinforcement Learning
Skill Learning: [C2]
[C5] FALCON: False-Negative Aware Learning of Contrastive Negatives in Vision-Language Alignment
Myunsoo Kim*, Seong-Woong Shim*, Byung-Jun Lee
Conference on Computer Vision and Pattern Recognition (CVPR 2026)
[C4] Beyond RAG vs. Long-Context: Learning Distraction-Aware Retrieval for Efficient Knowledge Grounding
Seong-Woong Shim*, Myunsoo Kim*, Jae Hyeon Cho, Byung-Jun Lee
International Conference on Learning Representations (ICLR 2026)
[C2] NBDI: A Simple and Effective Termination Condition for Skill Extraction from Task-Agnostic Demonstrations
Myunsoo Kim*, Hayeong Lee*, Seong-Woong Shim, JunHo Seo, Byung-Jun Lee
International Conference on Machine Learning (ICML 2025)
[C1] Adaptive Non-uniform Timestep Sampling for Accelerating Diffusion Model Training (KU Best Paper Award)
Myunsoo Kim*, Donghyeon Ki*, Seong-Woong Shim, Byung-Jun Lee
Conference on Computer Vision and Pattern Recognition (CVPR 2025)
Korea University, Seoul, Korea, Integrated MS. & Ph.D. (2023 - Present)
Artificial Intelligence Ph.D. candidate
GPA: 4.16/4.5
Korea University, Seoul, Korea, B.S. (2016 - 2023)
Statistics Bachelor's degree
GPA: 4.08/4.5
Korean (Native)
English (Professional working proficiency) (TOEFL iBT: 106/120)
Chinese (Full professional proficiency) (HSK 6級: 273/300)
Japanese (Professional working proficiency) (JLPT N1: 139/180)
TA, Advanced Language AI (LG CNS) (2023)
Conference on Neural Information Processing Systems (NeurIPS) 2026
Conference on Computer Vision and Pattern Recognition (CVPR) 2026
International Conference on Learning Representations (ICLR) 2026