Diffusion Based Unpaired Data Learning for Inverse Problems

arXiv:2609.01370v1 Announce Type: new Abstract: Data is important in many deep learning-based inverse problem solvers. However, obtaining sufficient paired data in many scenarios remains highly challenging, while unpaired data is cheap. To maximize data utilization, this paper proposes LUD-DIF, a diffusion-based approach for solving inverse problems with unpaired data. Starting from the evidence lower bound (ELBO) of the joint distribution, we decouple it into two independent diffusion processes under the weak-coupling assumption. The method provides theoretical support from a variational inference perspective, derives the loss function, quantitatively analyzes the error bound introduced by the assumption, and offers a theorem-motivated heuristic for hyperparameter selection. Experimental results demonstrate that LUD-DIF achieves outstanding performance on multiple image inverse problems, validating its effectiveness and generalization capability in unpaired inverse problem settings.

Related

• Measurement-Consistent Langevin Corrector for Stabilizing Latent Diffusion Inverse Problem Solvers
• Hallucination-Aware Diffusion Sampling for Inverse Problems via Robust Prior Updates
• Stochastic Optimal Control Sampling for Diffusion Inverse Problems
• MACS: Measurement-Aware Consistency Sampling for Inverse Problems

Source: arXiv cs.CV | 2026-09-02