mlsanityData to 7 Oct 2026

Training-Free Consistent Text-to-Image Generation

arXiv 2402.03286 ↗First posted 5 Feb 2024cs.CV cs.AI cs.GRcited by 123

Not on the map. Why?

Cited by 58 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Text-to-image models offer a new level of creative flexibility by allowing users to guide the image generation process through natural language. However, using these models to consistently portray the same subject across diverse prompts remains challenging. Existing approaches fine-tune the model to teach it new words that describe specific user-provided subjects or add image conditioning to the model. These methods require lengthy per-subject optimization or large-scale pre-training. Moreover, they struggle to align generated images with text prompts and face difficulties in portraying multiple subjects. Here, we present ConsiStory, a training-free approach that enables consistent subject generation by sharing the internal activations of the pretrained model. We introduce a subject-driven shared attention block and correspondence-based feature injection to promote subject consistency between images. Additionally, we develop strategies to encourage layout diversity while maintaining subject consistency. We compare ConsiStory to a range of baselines, and demonstrate state-of-the-art performance on subject consistency and text alignment, without requiring a single optimization step. Finally, ConsiStory can naturally extend to multi-subject scenarios, and even enable training-free personalization for common objects.

Key facts. Cited by 123 papers in the map (data through 7 Oct 2026). Builds on 42 of its 55 references that are in the map. First posted on arXiv 5 Feb 2024.

Builds on

42 of 55 references.

References · 42

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  3. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  4. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,388
  5. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
  6. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 1,998
  7. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,099
  8. 2023MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editingcited by 722
  9. 2023Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Modelscited by 696
  10. 2023Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animationcited by 683
  11. 2023Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generatorscited by 666
  12. 2023Zero-shot Image-to-Image Translationcited by 503
30 more
  1. 2023FateZero: Fusing Attentions for Zero-shot Text-based Video Editingcited by 462
  2. 2023TokenFlow: Consistent Diffusion Features for Consistent Video Editingcited by 440
  3. 2023BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editingcited by 438
  4. 2023Emergent Correspondence from Image Diffusioncited by 437
  5. 2023InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuningcited by 343
  6. 2023MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Modelcited by 330
  7. 2023Pix2Video: Video Editing using Image Diffusioncited by 318
  8. 2023A Tale of Two Features: Stable Diffusion Complements DINO for Zero-Shot Semantic Correspondencecited by 297
  9. 2023Mix-of-Show: Decentralized Low-Rank Adaptation for Multi-Concept Customization of Diffusion Modelscited by 261
  10. 2023An Edit Friendly DDPM Noise Space: Inversion and Manipulationscited by 243
  11. 2023Break-A-Scene: Extracting Multiple Concepts from a Single Imagecited by 230
  12. 2023Style Aligned Image Generation via Shared Attentioncited by 214
  13. 2023Diffusion Hyperfeatures: Searching Through Time and Space for Semantic Correspondencecited by 211
  14. 2022Splicing ViT Features for Semantic Appearance Transfercited by 209
  15. 2023Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Modelscited by 203
  16. 2023FreeU: Free Lunch in Diffusion U-Netcited by 195
  17. 2023Key-Locked Rank One Editing for Text-to-Image Personalizationcited by 194
  18. 2023DragonDiffusion: Enabling Drag-style Manipulation on Diffusion Modelscited by 193
  19. 2023Localizing Object-level Shape Variations with Text-to-Image Diffusion Modelscited by 142
  20. 2023Cross-Image Attention for Zero-Shot Appearance Transfercited by 135
  21. 2017Visual Attribute Transfer through Deep Image Analogycited by 123
  22. 2023Unsupervised Semantic Correspondence Using Stable Diffusioncited by 120
  23. 2023Domain-Agnostic Tuning-Encoder for Fast Personalization of Text-To-Image Modelscited by 85
  24. 2023TaleCrafter: Interactive Story Visualization with Multiple Characterscited by 71
  25. 2023Orthogonal Adaptation for Modular Customization of Diffusion Modelscited by 70
  26. 2023MotionEditor: Editing Video Motion via Content-Aware Diffusioncited by 56
  27. 2023SLiMe: Segment Like Mecited by 34
  28. 2023Improved Visual Story Generation with Adaptive Context Modelingcited by 22
  29. 2020Structural-analogy from a Single Image Paircited by 14
  30. 2023Contextual Knowledge Pursuit for Faithful Visual Synthesiscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 39; 2025: 63; 2026: 21.

123 papers.

Sort:

By area:

  1. 2024StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generationcited by 240
  2. 2024Lumina-T2X: Transforming Text into Any Modality, Resolution, and Duration via Flow-based Large Diffusion Transformerscited by 137
  3. 2025Emu3.5: Native Multimodal Models are World Learnerscited by 119
  4. 2024Stable Flow: Vital Layers for Training-Free Image Editingcited by 111
  5. 2024MovieDreamer: Hierarchical Generation for Coherent Long Visual Sequencecited by 85
  6. 2024SEED-Story: Multimodal Long Story Generation with Large Language Modelcited by 70
  7. 2025Multi-subject Open-set Personalization in Video Generationcited by 65
  8. 2024TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Modelscited by 64
  9. 2024MoA: Mixture-of-Attention for Subject-Context Disentanglement in Personalized Image Generationcited by 51
  10. 2024Large-Scale Text-to-Image Model with Inpainting is a Zero-Shot Subject-Driven Image Generatorcited by 48
  11. 2024Make It Count: Text-to-Image Generation with an Accurate Number of Objectscited by 47
  12. 2024DreamStory: Open-Domain Story Visualization by LLM-Guided Multi-Subject Consistent Diffusioncited by 46
38 more
  1. 2024Add-it: Training-Free Object Insertion in Images With Pretrained Diffusion Modelscited by 42
  2. 2024Looking Backward: Streaming Video-to-Video Translation with Feature Bankscited by 38
  3. 2025ShotAdapter: Text-to-Multi-Shot Video Generation with Diffusion Modelscited by 37
  4. 2025StoryMem: Multi-shot Long Video Storytelling with Memorycited by 35
  5. 2025EchoShot: Multi-Shot Portrait Video Generationcited by 28
  6. 2024StoryMaker: Towards Holistic Consistent Characters in Text-to-image Generationcited by 27
  7. 2025Generating Multi-Image Synthetic Data for Text-to-Image Customizationcited by 25
  8. 2024CharacterFactory: Sampling Consistent Characters with GANs for Diffusion Modelscited by 24
  9. 2024DiffUHaul: A Training-Free Method for Object Dragging in Imagescited by 22
  10. 2025Personalize Anything for Free with Diffusion Transformercited by 22
  11. 2025Nested Attention: Semantic-aware Attention Values for Concept Personalizationcited by 21
  12. 2024TheaterGen: Character Management with LLM for Consistent Multi-turn Image Generationcited by 18
  13. 2024Where's Waldo: Diffusion Features for Personalized Segmentation and Retrievalcited by 18
  14. 2025CharaConsist: Fine-Grained Consistent Character Generationcited by 18
  15. 2024From Parts to Whole: A Unified Reference Framework for Controllable Human Image Generationcited by 17
  16. 2025AnyStory: Towards Unified Single and Multiple Subject Personalization in Text-to-Image Generationcited by 16
  17. 2025Story2Board: A Training-Free Approach for Expressive Storyboard Generationcited by 16
  18. 2024OneActor: Consistent Character Generation via Cluster-Conditioned Guidancecited by 15
  19. 2025Cut2Next: Generating Next Shot via In-Context Tuningcited by 15
  20. 2024Unbounded: A Generative Infinite Game of Character Life Simulationcited by 14
  21. 2024InstantRestore: Single-Step Personalized Face Restoration with Shared-Image Attentioncited by 13
  22. 2025Training-free Geometric Image Editing on Diffusion Modelscited by 13
  23. 2024Story-Iter: A Training-free Iterative Paradigm for Long Story Visualizationcited by 12
  24. 2026ViMax: Agentic Video Generationcited by 11
  25. 2025IC-Custom: Diverse Image Customization via In-Context Learningcited by 10
  26. 2024Edicho: Consistent Image Editing in the Wildcited by 9
  27. 2025DynVFX: Augmenting Real Videos with Dynamic Contentcited by 9
  28. 2024Storynizor: Consistent Story Generation via Inter-Frame Synchronized and Shuffled ID Injectioncited by 8
  29. 2025EditID: Training-Free Editable ID Customization for Text-to-Image Generationcited by 8
  30. 2025MultiHuman-Testbench: Benchmarking Image Generation for Multiple Humanscited by 8
  31. 2025EditIDv2: Editable ID Customization with Data-Lubricated ID Feature Integration for Text-to-Image Generationcited by 8
  32. 2025Visual Persona: Foundation Model for Full-Body Human Customizationcited by 7
  33. 2025Object Isolated Attention for Consistent Story Visualizationcited by 7
  34. 2025In-Context Brush: Zero-shot Customized Subject Insertion with Context-Aware Latent Space Manipulationcited by 7
  35. 2025SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistencycited by 7
  36. 2025PaCo-RL: Advancing Reinforcement Learning for Consistent Image Generation with Pairwise Reward Modelingcited by 7
  37. 2024AttenCraft: Attention-guided Disentanglement of Multiple Concepts for Text-to-Image Customizationcited by 6
  38. 2024Character-Adapter: Prompt-Guided Region Control for High-Fidelity Character Customizationcited by 6

50 most cited of 123 All 123 (JSON)