Manifold-Optimal Guidance: A Unified Riemannian Control View of Diffusion Guidance
This addresses a key limitation in conditional diffusion models for AI image generation, offering a novel solution to improve output quality.
The paper tackled the problem of oversaturation and artifacts in diffusion models by proposing Manifold-Optimal Guidance (MOG), a geometry-aware control method that corrects off-manifold drift without retraining, resulting in superior fidelity and alignment with no added computational overhead.
Classifier-Free Guidance (CFG) serves as the de facto control mechanism for conditional diffusion, yet high guidance scales notoriously induce oversaturation, texture artifacts, and structural collapse. We attribute this failure to a geometric mismatch: standard CFG performs Euclidean extrapolation in ambient space, inadvertently driving sampling trajectories off the high-density data manifold. To resolve this, we present Manifold-Optimal Guidance (MOG), a framework that reformulates guidance as a local optimal control problem. MOG yields a closed-form, geometry-aware Riemannian update that corrects off-manifold drift without requiring retraining. Leveraging this perspective, we further introduce Auto-MOG, a dynamic energy-balancing schedule that adaptively calibrates guidance strength, effectively eliminating the need for manual hyperparameter tuning. Extensive validation demonstrates that MOG yields superior fidelity and alignment compared to baselines, with virtually no added computational overhead.