DAPS++: Rethinking Diffusion Inverse Problems with Decoupled Posterior Annealing

From a Bayesian perspective, score-based dix001Busion solves in-verse problems through joint inference, embedding the likelihood withthe prior to guide the sampling process. However, this formulation failsto explain its practical behavior: the prior ox001Bers limited guidance, whilereconstruction is largely driven by the measurement-consistency term,leading to an inference process that is ex001Bectively decoupled from thedix001Busion dynamics. We show that the dix001Busion prior in these solversfunctions primarily as a warm initializer that places estimates near thedata manifold, while reconstruction is driven almost entirely by measure-ment consistency. Based on this observation, we introduce DAPS++,which fully decouples dix001Busion-based initialization from likelihood-drivenrex001Cnement, allowing the likelihood term to guide inference more directlywhile maintaining numerical stability and providing insight into whyunix001Ced dix001Busion trajectories remain ex001Bective in practice. By requiringfewer function evaluations (NFEs) and measurement-optimization steps,DAPS++ achieves high computational ex001Eciency and robust reconstruc-tion performance across diverse image restoration tasks.