[2603.03710] MPFlow: Multi-modal Posterior-Guided Flow Matching for Zero-Shot MRI Reconstruction

Computer Science > Computer Vision and Pattern Recognition arXiv:2603.03710 (cs) [Submitted on 4 Mar 2026] Title:MPFlow: Multi-modal Posterior-Guided Flow Matching for Zero-Shot MRI Reconstruction Authors:Seunghoi Kim, Chen Jin, Henry F. J. Tregidgo, Matteo Figini, Daniel C. Alexander View a PDF of the paper titled MPFlow: Multi-modal Posterior-Guided Flow Matching for Zero-Shot MRI Reconstruction, by Seunghoi Kim and Chen Jin and Henry F. J. Tregidgo and Matteo Figini and Daniel C. Alexander View PDF HTML (experimental) Abstract:Zero-shot MRI reconstruction relies on generative priors, but single-modality unconditional priors produce hallucinations under severe ill-posedness. In many clinical workflows, complementary MRI acquisitions (e.g. high-quality structural scans) are routinely available, yet existing reconstruction methods lack mechanisms to leverage this additional information. We propose MPFlow, a zero-shot multi-modal reconstruction framework built on rectified flow that incorporates auxiliary MRI modalities at inference time without retraining the generative prior to improve anatomical fidelity. Cross-modal guidance is enabled by our proposed self-supervised pretraining strategy, Patch-level Multi-modal MR Image Pretraining (PAMRI), which learns shared representations across modalities. Sampling is jointly guided by data consistency and cross-modal feature alignment using pre-trained PAMRI, systematically suppressing intrinsic and extrinsic hallucinations. Exte...