paper-with-me

홈 › Papers

Hallucination-Aware Diffusion Sampling for Inverse Problems via Robust Prior Updates

2026-06-01 · Pengfei Jin, Yiqi Tian, Kailong Fan, Bingjie Qi, Quanzheng Li arxiv

Diffusion-based inverse problem solvers can produce realistic reconstructions, but realism alone does not ensure that the recovered details are supported by the measurement. We study this failure as measurement-conditioned hallucination: visually meaningful content that is either implausible or inconsistent with the measured instance. Our analysis separates Bayes-rule-based diffusion inverse solvers into a prior update and a measurement-conditioning step, showing that hallucinated content can enter through the prior-side proposal before the measurement correction is applied. Motivated by this view, we propose Robust Prior Update (RPU), a solver-level module that probes the local stability of the diffusion prior update, re-anchors the resulting displacement at the current iterate, and leaves the measurement update unchanged. We instantiate RPU in DPS and evaluate it on FFHQ and ImageNet inverse problems using automatic metrics and human faithfulness studies. On FFHQ, RPU improves PSNR and LPIPS over DPS across box inpainting, Gaussian deblurring, and motion deblurring. In human judgments, RPU receives 91.9% of blind non-tie majority preferences and 91.1% of ground-truth-assisted non-tie preferences on FFHQ box inpainting, while the ImageNet Gaussian reader study is tie-heavy but favors RPU among non-tie cases. These results support a targeted claim: robustifying the prior update can improve instance faithfulness in diffusion inverse solvers, especially when the prior shapes weakly constrained content.

📄 PDF Abstract BibTeX arXiv:2606.02331

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Decoupled Diffusion Sampling for Inverse Problems on Function Spaces

2026-01-30 · Thomas Y. L. Lin, Jiachen Yao, Lufang Chiang, Julius Berner 외 arxiv

We propose a data-efficient, physics-aware generative framework in function space for inverse PDE problems. Existing plug-and-play diffusion posterior samplers represent physics implicitly through joint coefficient-solut…

Diffusion Posterior Sampling for General Noisy Inverse Problems

2022-09-29 · Hyungjin Chung, Jeongsol Kim, Michael T. McCann, Marc L. Klasky 외

Diffusion models have been recently studied as powerful generative inverse problem solvers, owing to their high quality reconstructions and the ease of combining existing iterative solvers. However, most works focus on s…

DeblurringRetrieval

MACS: Measurement-Aware Consistency Sampling for Inverse Problems

2025-10-02 · Amirreza Tanevardi, Pooria Abbas Rad Moghadam, Seyed Mohammad Eshtehardian, Sajjad Amini 외 arxiv

Diffusion models have emerged as powerful generative priors for solving inverse imaging problems. However, their practical deployment is hindered by the substantial computational cost of slow, multi-step sampling. Althou…

Computational Efficiency

Solving Video Inverse Problems Using Image Diffusion Models

2024-09-04 · Taesung Kwon, Jong Chul Ye

Recently, diffusion model-based inverse problem solvers (DIS) have emerged as state-of-the-art approaches for addressing inverse problems, including image super-resolution, deblurring, inpainting, etc. However, their app…

DeblurringImage Super-ResolutionSuper-Resolution

Enhancing Diffusion Models for Inverse Problems with Covariance-Aware Posterior Sampling

2024-12-28 · Shayan Mohajer Hamidi, En-hui Yang

Inverse problems exist in many disciplines of science and engineering. In computer vision, for example, tasks such as inpainting, deblurring, and super resolution can be effectively modeled as inverse problems. Recently,…

DeblurringDenoisingSuper-Resolution