# Seg4Diff: Unveiling Open-Vocabulary Segmentation in Text-to-Image Diffusion Transformers

arXiv: https://arxiv.org/abs/2509.18096 · HTML: https://mlsanity.com/paper/2509.18096/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 63 of its 75 references that are in the map.
- First posted on arXiv 22 Sep 2025.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

## Builds on (most cited first, 30 of 63)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [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)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://mlsanity.com/paper/2107.06278/) (2021)
- [Semantic Understanding of Scenes through the ADE20K Dataset](https://mlsanity.com/paper/1608.05442/) (2016)
- [MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing](https://mlsanity.com/paper/2304.08465/) (2023)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [Extract Free Dense Labels from CLIP](https://mlsanity.com/paper/2112.01071/) (2021)
- [Emergent Correspondence from Image Diffusion](https://mlsanity.com/paper/2306.03881/) (2023)
- [Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.04803/) (2023)

## Built on by (most cited first, 24 of 24)

- [Emergent Outlier View Rejection in Visual Geometry Grounded Transformers](https://mlsanity.com/paper/2512.04012/) (2025)
- [C3G: Learning Compact 3D Representations with 2K Gaussians](https://mlsanity.com/paper/2512.04021/) (2025)
- [CAMEO: Correspondence-Attention Alignment for Multi-View Diffusion Models](https://mlsanity.com/paper/2512.03045/) (2025)
- [ScreenParse: Moving Beyond Sparse Grounding with Complete Screen Parsing Supervision](https://mlsanity.com/paper/2602.14276/) (2026)
- [Tiny Inference-Time Scaling with Latent Verifiers](https://mlsanity.com/paper/2603.22492/) (2026)
- [TORA: Topological Representation Alignment for 3D Shape Assembly](https://mlsanity.com/paper/2604.04050/) (2026)
- [Diffusion Model as a Generalist Segmentation Learner](https://mlsanity.com/paper/2604.24575/) (2026)
- [V-Warper: Appearance-Consistent Video Diffusion Personalization via Value Warping](https://mlsanity.com/paper/2512.12375/) (2025)
- [Entropy-Gradient Grounding: Training-Free Evidence Retrieval in Vision-Language Models](https://mlsanity.com/paper/2604.08456/) (2026)
- [Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers](https://mlsanity.com/paper/2605.14270/) (2026)
- [No Pose, No Problem in 4D: Feed-Forward Dynamic Gaussians from Unposed Multi-View Videos](https://mlsanity.com/paper/2605.22190/) (2026)
- [Interpretable Motion-Attentive Maps: Spatio-Temporally Localizing Concepts in Video Diffusion Transformers](https://mlsanity.com/paper/2603.02919/) (2026)
- [Unlocking Diffusion Hierarchies: Adaptive Timestep Selection for Zero-Shot Segmentation](https://mlsanity.com/paper/2606.15590/) (2026)
- [Exploring Conditions for Diffusion models in Robotic Control](https://mlsanity.com/paper/2510.15510/) (2025)
- [EraseLoRA: MLLM-Driven Foreground Exclusion and Background Subtype Aggregation for Dataset-Free Object Removal](https://mlsanity.com/paper/2512.21545/) (2025)
- [MagicFuse: Single Image Fusion for Visual and Semantic Reinforcement](https://mlsanity.com/paper/2602.01760/) (2026)
- [The Latent Color Subspace: Emergent Order in High-Dimensional Chaos](https://mlsanity.com/paper/2603.12261/) (2026)
- [DA-Flow: Degradation-Aware Optical Flow Estimation with Diffusion Models](https://mlsanity.com/paper/2603.23499/) (2026)
- [Spectral Forensics of Diffusion Attention Graphs for Copy-Move Forgery Detection](https://mlsanity.com/paper/2604.17287/) (2026)
- [PROSE: Training-Free Egocentric Scene Registration with Vision-Language Models](https://mlsanity.com/paper/2606.16569/) (2026)
- [MMDiff: Extending Diffusion Transformers for Multi-Modal Generation](https://mlsanity.com/paper/2606.16673/) (2026)
- [Controlling Motion Transfer in Diffusion Transformers via Attention Heads](https://mlsanity.com/paper/2607.11081/) (2026)
- [MAVISEG: Manifold Propagation and Visual Prototypes for Zero-Shot Open-Vocabulary Segmentation in Diffusion Transformers](https://mlsanity.com/paper/2608.05878/) (2026)
- [Diffuse2Seg: Diffusion Models Can Segment Anything Without Supervision](https://mlsanity.com/paper/2609.06491/) (2026)

## Abstract (arXiv)

Text-to-image diffusion models excel at translating language prompts into photorealistic images by implicitly grounding textual concepts through their cross-modal attention mechanisms. Recent multi-modal diffusion transformers extend this by introducing joint self-attention over concatenated image and text tokens, enabling richer and more scalable cross-modal alignment. However, a detailed understanding of how and where these attention maps contribute to image generation remains limited. In this paper, we introduce Seg4Diff (Segmentation for Diffusion), a systematic framework for analyzing the attention structures of MM-DiT, with a focus on how specific layers propagate semantic information from text to image. Through comprehensive analysis, we identify a semantic grounding expert layer, a specific MM-DiT block that consistently aligns text tokens with spatially coherent image regions, naturally producing high-quality semantic segmentation masks. We further demonstrate that applying a lightweight fine-tuning scheme with mask-annotated image data enhances the semantic grouping capabilities of these layers and thereby improves both segmentation performance and generated image fidelity. Our findings demonstrate that semantic grouping is an emergent property of diffusion transformers and can be selectively amplified to advance both segmentation and generation performance, paving the way for unified models that bridge visual perception and generation.