# MiM-DiT: MoE in MoE with Diffusion Transformers for All-in-One Image Restoration

arXiv: https://arxiv.org/abs/2603.02710 · HTML: https://mlsanity.com/paper/2603.02710/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 4 of its 50 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Placed in

- Diffusion models / [Inverse problems and image restoration](https://mlsanity.com/diffusion/inverse-problems-and-image-restoration/)
- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 4 of 4)

- [Controlling Vision-Language Models for Multi-Task Image Restoration](https://mlsanity.com/paper/2310.01018/) (2023)
- [ProRes: Exploring Degradation-aware Visual Prompt for Universal Image Restoration](https://mlsanity.com/paper/2306.13653/) (2023)
- [Toward Real-world Single Image Deraining: A New Benchmark and Beyond](https://mlsanity.com/paper/2206.05514/) (2022)
- [UniRestorer: Universal Image Restoration via Adaptively Estimating Image Degradation at Proper Granularity](https://mlsanity.com/paper/2412.20157/) (2024)

## Abstract (arXiv)

All-in-one image restoration is challenging because different degradation types, such as haze, blur, noise, and low-light, impose diverse requirements on restoration strategies, making it difficult for a single model to handle them effectively. In this paper, we propose a unified image restoration framework that integrates a dual-level Mixture-of-Experts (MoE) architecture with a pretrained diffusion model. The framework operates at two levels: the Inter-MoE layer adaptively combines expert groups to handle major degradation types, while the Intra-MoE layer further selects specialized sub-experts to address fine-grained variations within each type. This design enables the model to achieve coarse-grained adaptation across diverse degradation categories while performing fine-grained modulation for specific intra-class variations, ensuring both high specialization in handling complex, real-world corruptions. Extensive experiments demonstrate that the proposed method performs favorably against the state-of-the-art approaches on multiple image restoration task.