Adding Additional Control to One-Step Diffusion with Joint Distribution Matching
By: Yihong Luo , Tianyang Hu , Yifan Song and more
Potential Business Impact:
Makes AI create pictures from new ideas faster.
While diffusion distillation has enabled one-step generation through methods like Variational Score Distillation, adapting distilled models to emerging new controls -- such as novel structural constraints or latest user preferences -- remains challenging. Conventional approaches typically requires modifying the base diffusion model and redistilling it -- a process that is both computationally intensive and time-consuming. To address these challenges, we introduce Joint Distribution Matching (JDM), a novel approach that minimizes the reverse KL divergence between image-condition joint distributions. By deriving a tractable upper bound, JDM decouples fidelity learning from condition learning. This asymmetric distillation scheme enables our one-step student to handle controls unknown to the teacher model and facilitates improved classifier-free guidance (CFG) usage and seamless integration of human feedback learning (HFL). Experimental results demonstrate that JDM surpasses baseline methods such as multi-step ControlNet by mere one-step in most cases, while achieving state-of-the-art performance in one-step text-to-image synthesis through improved usage of CFG or HFL integration.
Similar Papers
Learning Few-Step Diffusion Models by Trajectory Distribution Matching
CV and Pattern Recognition
Makes AI art and video creation much faster.
Adversarial Distribution Matching for Diffusion Distillation Towards Efficient Image and Video Synthesis
CV and Pattern Recognition
Makes AI create better pictures and videos faster.
Distillation of Discrete Diffusion by Exact Conditional Distribution Matching
Machine Learning (CS)
Makes AI create pictures much faster.