|
Shikun Sun I am a third-year PhD candidate in the Department of Computer Science and Technology (DCST) at Tsinghua University, under the supervision of Professor Jia Jia. I earned my B.Eng. degree from the same department in 2021. Additionally, I received a second B.Eng. degree from the Department of Mathematical Sciences at Tsinghua University in 2021, where I worked with Professor Chenglong Bao. |
|
Selected ResearchMy primary research focus is on diffusion models, with a specific interest in advancing new theoretical frameworks, enhancing controllable generation, exploring RL-based fine-tuning techniques, and improving computational efficiency. |
|
Inner Classifier-Free Guidance and Its Taylor Expansion for Diffusion Models
Shikun Sun, Longhui Wei, Zhicai Wang, Zixuan Wang, Junliang Xing, Jia Jia, Qi Tian ICLR, 2024 Paper / OpenReview Inner classifier-free guidance (ICFG) provides an alternative perspective on the CFG method when the condition has a specific structure, demonstrating that CFG represents a first-order case of ICFG. |
|
SDDM: Score-Decomposed Diffusion Models on Manifolds for Unpaired Image-to-Image Translation.
Shikun Sun, Longhui Wei, Junliang Xing, Jia Jia, Qi Tian ICML, 2023 Paper / arXiv This work presents a new score-decomposed diffusion model (SDDM) on manifolds to explicitly optimize the tangled distributions during image generation. |
Miscellanea |
|
Based on Jon Barron's website. |