8 papers
Breaking the Curse of Dimensionality: Diffusion Models Efficiently Learn Low-Dimensional Distributions
Peng Wang, Huijie Zhang, Zekai Zhang +3
Despite their empirical success across a wide range of generative tasks, the fundamental principles underlying the ability of diffusion models to learn data distributions are poorl…
The Emergence of Reproducibility and Generalizability in Diffusion Models
Huijie Zhang, Jinfan Zhou, Yifu Lu +4
In this work, we investigate an intriguing and prevalent phenomenon of diffusion models which we term as "consistent model reproducibility": given the same starting noise input and…
Exploring Low-Dimensional Subspaces in Diffusion Models for Controllable Image Editing
Siyi Chen, Huijie Zhang, Minzhe Guo +3
Recently, diffusion models have emerged as a powerful class of generative models. Despite their success, there is still limited understanding of their semantic spaces. This makes i…
Understanding Generalization in Diffusion Distillation via Probability Flow Distance
Huijie Zhang, Zijian Huang, Siyi Chen +4
Diffusion distillation provides an effective approach for learning lightweight and few-steps diffusion models with efficient generation. However, evaluating their generalization re…
Shallow Diffuse: Robust and Invisible Watermarking through Low-Dimensional Subspaces in Diffusion Models
Wenda Li, Huijie Zhang, Qing Qu
The widespread use of AI-generated content from diffusion models has raised significant concerns regarding misinformation and copyright infringement. Watermarking is a crucial tech…
Improving Efficiency of Diffusion Models via Multi-Stage Framework and Tailored Multi-Decoder Architectures
Huijie Zhang, Yifu Lu, Ismail Alkhouri +3
Diffusion models, emerging as powerful deep generative tools, excel in various applications. They operate through a two-steps process: introducing noise into training samples and t…