mlsanityData to 14 Oct 2026

Stable Flow: Vital Layers for Training-Free Image Editing

arXiv 2411.14430 ↗First posted 21 Nov 2024cs.CV cs.GR cs.LGcited by 113

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

Diffusion models have revolutionized the field of content synthesis and editing. Recent models have replaced the traditional UNet architecture with the Diffusion Transformer (DiT), and employed flow-matching for improved training and sampling. However, they exhibit limited generation diversity. In this work, we leverage this limitation to perform consistent image edits via selective injection of attention features. The main challenge is that, unlike the UNet-based models, DiT lacks a coarse-to-fine synthesis structure, making it unclear in which layers to perform the injection. Therefore, we propose an automatic method to identify "vital layers" within DiT, crucial for image formation, and demonstrate how these layers facilitate a range of controlled stable edits, from non-rigid modifications to object addition, using the same mechanism. Next, to enable real-image editing, we introduce an improved image inversion method for flow models. Finally, we evaluate our approach through qualitative and quantitative comparisons, along with a user study, and demonstrate its effectiveness across multiple applications. The project page is available at https://omriavrahami.com/stable-flow

Key facts. Cited by 113 papers in the map (data through 14 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 92 of its 102 references that are in the map. First posted on arXiv 21 Nov 2024.

Builds on

92 of 102 references.

Closest ancestors · 29

Same direction, newest first

  1. 2024TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models1 Aug
  2. 2024DiffUHaul: A Training-Free Method for Object Dragging in Images3 Jun
  3. 2024Lazy Diffusion Transformer for Interactive Image Editing18 Apr
  4. 2024Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation25 Mar
  5. 2024ReNoise: Real Image Inversion Through Iterative Noising21 Mar
  6. 2024Diffusion Model-Based Image Editing: A Survey27 Feb
  7. 2024Visual Style Prompting with Swapping Self-Attention20 Feb
  8. 2023ZONE: Zero-Shot Instruction-Guided Local Editing28 Dec
  9. 2023LEDITS++: Limitless Image Editing using Text-to-Image Models28 Nov
  10. 2023The Chosen One: Consistent Characters in Text-to-Image Diffusion Models16 Nov
  11. 2023Forgedit: Text Guided Image Editing via Learning and Forgetting19 Sep
  12. 2023Effective Real Image Editing with Accelerated Iterative Diffusion Inversion10 Sep
17 more
  1. 2023Improving Tuning-Free Real Image Editing with Proximal Guidance8 Jun
  2. 2023InstructEdit: Improving Automatic Masks for Diffusion-based Image Editing With User Instructions29 May
  3. 2023An Edit Friendly DDPM Noise Space: Inversion and Manipulations12 Apr
  4. 2023Localizing Object-level Shape Variations with Text-to-Image Diffusion Models20 Mar
  5. 2023Region-Aware Diffusion for Zero-shot Text-driven Image Editing23 Feb
  6. 2022Imagen Editor and EditBench: Advancing and Evaluating Text-Guided Image Inpainting13 Dec
  7. 2022SpaText: Spatio-Textual Representation for Controllable Image Generation25 Nov
  8. 2022Paint by Example: Exemplar-based Image Editing with Diffusion Models23 Nov
  9. 2022InstructPix2Pix: Learning to Follow Image Editing Instructions17 Nov
  10. 2022Null-text Inversion for Editing Real Images using Guided Diffusion Models17 Nov
  11. 2022DiffEdit: Diffusion-based semantic image editing with mask guidance20 Oct
  12. 2022Imagic: Text-Based Real Image Editing with Diffusion Models17 Oct
  13. 2022Blended Latent Diffusion6 Jun
  14. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding23 May
  15. 2022Hierarchical Text-Conditional Image Generation with CLIP Latents13 Apr
  16. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models20 Dec
  17. 2021Blended Diffusion for Text-driven Editing of Natural Images29 Nov

Other references · 63

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  5. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  6. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  7. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  8. 2020Denoising Diffusion Implicit Modelscited by 9,781
  9. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  10. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  11. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  12. 2022Scalable Diffusion Models with Transformerscited by 6,054
38 more
  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  2. 2022Flow Matching for Generative Modelingcited by 5,702
  3. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
  4. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  5. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  6. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  7. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  8. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  9. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,431
  10. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  11. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  12. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,937
  13. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,917
  14. 2022Building Normalizing Flows with Stochastic Interpolantscited by 995
  15. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 991
  16. 2022Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generationcited by 961
  17. 2022Plug-and-Play Diffusion Features for Text-Driven Image-to-Image Translationcited by 906
  18. 2020Encoding in Style: a StyleGAN Encoder for Image-to-Image Translationcited by 727
  19. 2023MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editingcited by 727
  20. 2019Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?cited by 675
  21. 2023MagicBrush: A Manually Annotated Dataset for Instruction-Guided Image Editingcited by 624
  22. 2021Designing an Encoder for StyleGAN Image Manipulationcited by 544
  23. 2023Zero-shot Image-to-Image Translationcited by 505
  24. 2023TokenFlow: Consistent Diffusion Features for Consistent Video Editingcited by 442
  25. 2021Pivotal Tuning for Latent-based Editing of Real Imagescited by 427
  26. 2020In-Domain GAN Inversion for Real Image Editingcited by 420
  27. 2021GAN Inversion: A Surveycited by 367
  28. 2019Image2StyleGAN++: How to Edit the Embedded Images?cited by 356
  29. 2023Emu Edit: Precise Image Editing via Recognition and Generation Taskscited by 324
  30. 2022Text2LIVE: Text-Driven Layered Image and Video Editingcited by 312
  31. 2022SmartBrush: Text and Shape Guided Object Inpainting with Diffusion Modelcited by 278
  32. 2022EDICT: Exact Diffusion Inversion via Coupled Transformationscited by 258
  33. 2023Style Aligned Image Generation via Shared Attentioncited by 216
  34. 2021HyperStyle: StyleGAN Inversion with HyperNetworks for Real Image Editingcited by 208
  35. 2020Semantic Photo Manipulation with a Generative Image Priorcited by 178
  36. 2020Adversarial Latent Autoencoderscited by 155
  37. 2023Cross-Image Attention for Zero-Shot Appearance Transfercited by 136
  38. 2024Implicit Style-Content Separation using B-LoRAcited by 135

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 52; 2026: 60.

113 papers.

Sort:

By area:

  1. 2024FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Modelscited by 217
  2. 2025In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformercited by 204
  3. 2025Pico-Banana-400K: A Large-Scale Dataset for Text-Guided Image Editingcited by 61
  4. 2025KV-Edit: Training-Free Image Editing for Precise Background Preservationcited by 49
  5. 2025UniEdit-Flow: Unleashing Inversion and Editing in the Era of Flow Modelscited by 39
  6. 2025Stochastic Self-Guidance for Training-Free Enhancement of Diffusion Modelscited by 36
  7. 2025IMAGHarmony: Controllable Image Editing with Consistent Object Quantity and Layoutcited by 28
  8. 2025Follow-Your-Shape: Shape-Aware Image Editing via Trajectory-Guided Region Controlcited by 28
  9. 2025Seg4Diff: Unveiling Open-Vocabulary Segmentation in Text-to-Image Diffusion Transformerscited by 24
  10. 2025Edit Transfer: Learning Image Editing via Vision In-Context Relationscited by 23
  11. 2025FreeFlux: Understanding and Exploiting Layer-Specific Roles in RoPE-Based MMDiT for Versatile Image Editingcited by 22
  12. 2025Personalized Image Generation with Deep Generative Models: A Decade Surveycited by 20
38 more
  1. 2025Multi-turn Consistent Image Editingcited by 18
  2. 2025Exploring Multimodal Diffusion Transformers for Enhanced Prompt-based Image Editingcited by 17
  3. 2025Token Perturbation Guidance for Diffusion Modelscited by 15
  4. 2025SliderEdit: Continuous Image Editing with Fine-Grained Instruction Controlcited by 14
  5. 2025Adams Bashforth Moulton Solver for Inversion and Editing in Rectified Flowcited by 10
  6. 2025LayerPeeler: Autoregressive Peeling for Layer-wise Image Vectorizationcited by 8
  7. 2025Group Relative Attention Guidance for Image Editingcited by 8
  8. 2026From Scale to Speed: Adaptive Test-Time Scaling for Image Editingcited by 8
  9. 2025PartEdit: Fine-Grained Image Editing using Pre-Trained Diffusion Modelscited by 7
  10. 2025Localizing Knowledge in Diffusion Transformerscited by 7
  11. 2025EditInfinity: Image Editing with Binary-Quantized Generative Modelscited by 7
  12. 2025Temporal Concept Dynamics in Diffusion Models via Prompt-Conditioned Interventionscited by 7
  13. 2025Where and How to Perturb: On the Design of Perturbation Guidance in Diffusion and Flow Modelscited by 6
  14. 2025FreqEdit: Preserving High-Frequency Features for Robust Multi-Turn Image Editingcited by 6
  15. 2025ProEdit: Inversion-based Editing From Prompts Done Rightcited by 6
  16. 2025DCI: Dual-Conditional Inversion for Boosting Diffusion-Based Image Editingcited by 5
  17. 2025FastFLUX: Pruning FLUX with Block-wise Replacement and Sandwich Trainingcited by 5
  18. 2025Taming Flow-based I2V Models for Creative Video Editingcited by 5
  19. 2025LaTo: Landmark-tokenized Diffusion Transformer for Fine-grained Human Face Editingcited by 5
  20. 2025TempoControl: Temporal Attention Guidance for Text-to-Video Modelscited by 5
  21. 2025SplitFlux: Learning to Decouple Content and Style from a Single Imagecited by 5
  22. 2026Continuous Control of Editing Models via Adaptive-Origin Guidancecited by 5
  23. 2026Layer-wise Instance Binding for Regional and Occlusion Control in Text-to-Image Diffusion Transformerscited by 5
  24. 2026Improving Robotic Generalist Policies via Flow Reversal Steeringcited by 5
  25. 2025Inverse Virtual Try-On: Generating Multi-Category Product-Style Images from Clothed Individualscited by 4
  26. 2025Rare Text Semantics Were Always There in Your Diffusion Transformercited by 4
  27. 2025Are Image-to-Video Models Good Zero-Shot Image Editors?cited by 4
  28. 2026TP-Blend: Textual-Prompt Attention Pairing for Precise Object-Style Blending in Diffusion Modelscited by 4
  29. 2026DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformerscited by 4
  30. 2026SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generationcited by 4
  31. 2025TV-LiVE: Training-Free, Text-Guided Video Editing via Layer Informed Vitality Exploitationcited by 3
  32. 2025FreeCus: Free Lunch Subject-driven Customization in Diffusion Transformerscited by 3
  33. 2025Consistent and Controllable Image Animation with Motion Linear Diffusion Transformerscited by 3
  34. 2025TweezeEdit: Consistent and Efficient Image Editing with Path Regularizationcited by 3
  35. 2025OmniText: A Training-Free Generalist for Controllable Text-Image Manipulationcited by 3
  36. 2026TexTailor: Inference-Time Textual Guidance Tailoring for Multimodal Diffusion Transformerscited by 3
  37. 2026On-the-fly Repulsion in the Contextual Space for Rich Diversity in Diffusion Transformerscited by 3
  38. 2026SteerFlow: Steering Rectified Flows for Faithful Inversion-Based Image Editingcited by 3

50 most cited of 113 All 113 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026AdaEdit: Adaptive Temporal and Channel Modulation for Flow-Based Image Editing
  2. 2024Streamlining Image Editing with Layered Diffusion Brushes
  3. 2024Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing
  4. 2026Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers
  5. 2024Unveil Inversion and Invariance in Flow Transformer for Versatile Image Editing
  6. 2026AttnRouter: Per-Category Attention Routing for Training-Free Image Editing on MMDiT
  7. 2025Self-Attention Decomposition For Training Free Diffusion Editing
  8. 2025Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers