Robust Posterior Diffusion-based Sampling via Adaptive Guidance Scale
This work addresses a key bottleneck in applying diffusion models to inverse problems in imaging, offering a robust and hyperparameter-free method that enhances reconstruction quality without task-specific tuning, though it is incremental as it builds on existing diffusion-based approaches.
The paper tackled the challenge of balancing prior and data fidelity in diffusion models for inverse problems by proposing an adaptive likelihood step-size strategy, resulting in improved reconstruction quality across imaging tasks like super-resolution and deblurring on datasets such as CelebA-HQ and ImageNet-256, with consistent gains in perceptual quality and minimal distortion loss.
Diffusion models have recently emerged as powerful generative priors for solving inverse problems, achieving state-of-the-art results across various imaging tasks. A central challenge in this setting lies in balancing the contribution of the prior with the data fidelity term: overly aggressive likelihood updates may introduce artifacts, while conservative updates can slow convergence or yield suboptimal reconstructions. In this work, we propose an adaptive likelihood step-size strategy to guide the diffusion process for inverse-problem formulations. Specifically, we develop an observation-dependent weighting scheme based on the agreement between two different approximations of the intractable intermediate likelihood gradients, that adapts naturally to the diffusion schedule, time re-spacing, and injected stochasticity. The resulting approach, Adaptive Posterior diffusion Sampling (AdaPS), is hyperparameter-free and improves reconstruction quality across diverse imaging tasks - including super-resolution, Gaussian deblurring, and motion deblurring - on CelebA-HQ and ImageNet-256 validation sets. AdaPS consistently surpasses existing diffusion-based baselines in perceptual quality with minimal or no loss in distortion, without any task-specific tuning. Extensive ablation studies further demonstrate its robustness to the number of diffusion steps, observation noise levels, and varying stochasticity.