# MMDiff: Extending Diffusion Transformers for Multi-Modal Generation

arXiv: https://arxiv.org/abs/2606.16673 · HTML: https://mlsanity.com/paper/2606.16673/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 45 of its 57 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 30 of 45)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Going deeper with Image Transformers](https://mlsanity.com/paper/2103.17239/) (2021)
- [Structure-measure: A New Way to Evaluate Foreground Maps](https://mlsanity.com/paper/1708.00786/) (2017)
- [Emergent Correspondence from Image Diffusion](https://mlsanity.com/paper/2306.03881/) (2023)
- [Unleashing Text-to-Image Diffusion Models for Visual Perception](https://mlsanity.com/paper/2303.02153/) (2023)
- [Diffusion Hyperfeatures: Searching Through Time and Space for Semantic Correspondence](https://mlsanity.com/paper/2305.14334/) (2023)
- [DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Models](https://mlsanity.com/paper/2303.11681/) (2023)
- [DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion Models](https://mlsanity.com/paper/2308.06160/) (2023)
- [DepthFormer: Exploiting Long-Range Correlation and Local Information for Accurate Monocular Depth Estimation](https://mlsanity.com/paper/2203.14211/) (2022)
- [Dataset Diffusion: Diffusion-based Synthetic Dataset Generation for Pixel-Level Semantic Segmentation](https://mlsanity.com/paper/2309.14303/) (2023)
- [Wonderland: Navigating 3D Scenes from a Single Image](https://mlsanity.com/paper/2412.12091/) (2024)
- [What Matters When Repurposing Diffusion Models for General Dense Perception Tasks?](https://mlsanity.com/paper/2403.06090/) (2024)
- [World-consistent Video Diffusion with Explicit 3D Modeling](https://mlsanity.com/paper/2412.01821/) (2024)

## Abstract (arXiv)

Diffusion transformers have demonstrated remarkable generative capabilities, yet the rich perceptual representations computed across their denoising trajectory are discarded once the content is rendered. We present MMDiff, a framework that transforms a frozen diffusion transformer into a multi-modal generative system that jointly produces images alongside any combination of dense perceptual modalities using lightweight decoder heads. Our central finding is that perceptual information is temporally distributed along the denoising trajectory, and that multi-timestep feature fusion with spatially varying aggregation weights is essential, improving semantic segmentation results by up to 28.7% mIoU over single-timestep extraction. We further adopt concept-driven attention extraction for interpretable spatial guidance, and show that frozen diffusion features are competitive with and complementary to state-of-the-art encoders such as DINOv3. By training only lightweight decoder heads on a frozen backbone, we achieve strong performance in semantic segmentation, salient object detection, and depth estimation, and demonstrate that this framework enables effective synthetic data generation at scale.