Direct Conditional Transition Sampling for Diffusion Inverse Problems

arXiv:2609.14596v1 Announce Type: new Abstract: Training-free diffusion inverse solvers typically choose between local measurement guidance and costly clean-space posterior updates. Independent posterior refresh can improve global correction by sampling a clean conditional and re-noising it, but its practical realization requires probability-flow ODE integration and clean-space Markov chain Monte Carlo (MCMC). We propose Direct Conditional Transition Sampling (DCTS), a direct stochastic-flow approximation to the same ideal refresh target. Rather than explicitly drawing a clean sample, DCTS estimates the measurement-conditioned clean mean along a short inner path and transports Gaussian source noise directly to the next noisy state. A denoiser-compatible sufficient statistic and a covariance-scaled operator update enable this conditional-mean estimation. Experiments on four inverse problems demonstrate that DCTS achieves competitive reconstruction quality with up to 16.8imes speedups over competing methods.

Related

• Accelerating Diffusion Sampling via Exploiting Local Transition Coherence
• MACS: Measurement-Aware Consistency Sampling for Inverse Problems
• Measurement-Consistent Langevin Corrector for Stabilizing Latent Diffusion Inverse Problem Solvers
• Discrete Langevin-Inspired Posterior Sampling

Source: arXiv cs.CV | 2026-09-15