# MARR: Decoupling Policy, Execution, and Calibration for All-in-One Medical Image Restoration

arXiv: https://arxiv.org/abs/2609.06645 · HTML: https://mlsanity.com/paper/2609.06645/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 34 references that are in the map.
- First posted on arXiv 6 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [SwinIR: Image Restoration Using Swin Transformer](https://mlsanity.com/paper/2108.10257/) (2021)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model](https://mlsanity.com/paper/2401.09417/) (2024)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [Simple Baselines for Image Restoration](https://mlsanity.com/paper/2204.04676/) (2022)
- [MambaIR: A Simple Baseline for Image Restoration with State-Space Model](https://mlsanity.com/paper/2402.15648/) (2024)
- [MambaIRv2: Attentive State Space Restoration](https://mlsanity.com/paper/2411.15269/) (2024)
- [Restore-RWKV: Efficient and Effective Medical Image Restoration with RWKV](https://mlsanity.com/paper/2407.11087/) (2024)
- [All-In-One Medical Image Restoration via Task-Adaptive Routing](https://mlsanity.com/paper/2405.19769/) (2024)
- [3D Segmentation Guided Style-based Generative Adversarial Networks for PET Synthesis](https://mlsanity.com/paper/2205.08887/) (2022)
- [DenoMamba: A fused state-space model for low-dose CT denoising](https://mlsanity.com/paper/2409.13094/) (2024)
- [DRMC: A Generalist Model with Dynamic Routing for Multi-Center PET Image Synthesis](https://mlsanity.com/paper/2307.05249/) (2023)
- [Learning Domain-Aware Task Prompt Representations for Multi-Domain All-in-One Image Restoration](https://mlsanity.com/paper/2603.01725/) (2026)
- [Region Attention Transformer for Medical Image Restoration](https://mlsanity.com/paper/2407.09268/) (2024)
- [All-in-One Medical Image Restoration with Latent Diffusion-Enhanced Vector-Quantized Codebook Prior](https://mlsanity.com/paper/2507.19874/) (2025)
- [QualiTeacher: Quality-Conditioned Pseudo-Labeling for Real-World Image Restoration](https://mlsanity.com/paper/2603.08030/) (2026)
- [TAT: Task-Adaptive Transformer for All-in-One Medical Image Restoration](https://mlsanity.com/paper/2512.14550/) (2025)
- [MAP-Diff: Multi-Anchor Guided Diffusion for Progressive 3D Whole-Body Low-Dose PET Denoising](https://mlsanity.com/paper/2603.02012/) (2026)

## Abstract (arXiv)

All-in-one medical image restoration seeks to recover heterogeneous clinical images with a single model, but PET, CT, and MRI differ substantially in degradation statistics, anatomical contrast, and output-space bias. A fully shared network can entangle modality-specific residual errors, whereas separate modality-specific networks sacrifice the practical advantages of unified deployment. We therefore recast all-in-one restoration as a question of where limited adaptation should be placed: policy selection, feature execution, or output calibration. We propose MARR, a compact restoration framework that constrains multi-modality adaptation into degradation-aware policy routing, modality-private residual execution, and image-domain residual correction without requiring degradation labels or separate modality-specific models. The policy branch forms a routing prompt from input statistics, latent content, and modality identity, and uses it only as a control signal. Prompt-gated modality-private adapters then perform lightweight residual refinement at intermediate decoder stages, while zero-initialized modality-specific output heads calibrate the final image-domain residual without perturbing the initial shared prediction. On an all-in-one PET, CT, and MRI restoration benchmark, MARR outperforms thirteen methods re-trained under the same protocol, achieving PSNR values of 37.34 dB, 33.85 dB, and 32.09 dB on PET, CT, and MRI, respectively, and the best modality-average PSNR of 34.43 dB. The code is publicly available at https://github.com/CHB-learner/MARR.