Variational Control for Guidance in Diffusion Models
This addresses the need for flexible and efficient guidance in diffusion models for tasks like inverse problems, though it appears incremental as it builds on existing variational and control perspectives.
The paper tackles the problem of guiding pretrained diffusion models without additional training, introducing Diffusion Trajectory Matching (DTM) that unifies guidance methods and achieves state-of-the-art results on various inverse problems.
Diffusion models exhibit excellent sample quality, but existing guidance methods often require additional model training or are limited to specific tasks. We revisit guidance in diffusion models from the perspective of variational inference and control, introducing Diffusion Trajectory Matching (DTM) that enables guiding pretrained diffusion trajectories to satisfy a terminal cost. DTM unifies a broad class of guidance methods and enables novel instantiations. We introduce a new method within this framework that achieves state-of-the-art results on several linear, non-linear, and blind inverse problems without requiring additional model training or specificity to pixel or latent space diffusion models. Our code will be available at https://github.com/czi-ai/oc-guidance