mlsanityData to 14 Oct 2026

Less-to-More Generalization: Unlocking More Controllability by In-Context Generation

arXiv 2504.02160 ↗First posted 2 Apr 2025cs.CV cs.LGcited by 154

Not on the map. Why?

Cited by 40 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Although subject-driven generation has been extensively explored in image generation due to its wide applications, it still has challenges in data scalability and subject expansibility. For the first challenge, moving from curating single-subject datasets to multiple-subject ones and scaling them is particularly difficult. For the second, most recent methods center on single-subject generation, making it hard to apply when dealing with multi-subject scenarios. In this study, we propose a highly-consistent data synthesis pipeline to tackle this challenge. This pipeline harnesses the intrinsic in-context generation capabilities of diffusion transformers and generates high-consistency multi-subject paired data. Additionally, we introduce UNO, which consists of progressive cross-modal alignment and universal rotary position embedding. It is a multi-image conditioned subject-to-image model iteratively trained from a text-to-image model. Extensive experiments show that our method can achieve high consistency while ensuring controllability in both single-subject and multi-subject driven generation.

Key facts. Cited by 154 papers in the map (data through 14 Oct 2026). Builds on 46 of its 52 references that are in the map. First posted on arXiv 2 Apr 2025.

Builds on

46 of 52 references.

References · 46

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  5. 2022Training language models to follow instructions with human feedbackcited by 15,971
  6. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  7. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  8. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  9. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  10. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  11. 2022Scalable Diffusion Models with Transformerscited by 6,054
  12. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
34 more
  1. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  2. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  4. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  5. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  6. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,245
  7. 2023IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Modelscited by 1,464
  8. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  9. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,106
  10. 2016Generative Adversarial Text to Image Synthesiscited by 735
  11. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 694
  12. 2022Improving alignment of dialogue agents via targeted human judgementscited by 531
  13. 2023Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervisioncited by 467
  14. 2023BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editingcited by 442
  15. 2024InstantID: Zero-shot Identity-Preserving Generation in Secondscited by 436
  16. 2023ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generationcited by 416
  17. 2024OmniGen: Unified Image Generationcited by 394
  18. 2022Self-critiquing models for assisting human evaluatorscited by 380
  19. 2024OminiControl: Minimal and Universal Control for Diffusion Transformercited by 340
  20. 2023MVImgNet: A Large-scale Dataset of Multi-view Imagescited by 220
  21. 2022Re-Imagen: Retrieval-Augmented Text-to-Image Generatorcited by 210
  22. 2023Subject-Diffusion:Open Domain Personalized Text-to-Image Generation without Test-time Fine-tuningcited by 191
  23. 2024In-Context LoRA for Diffusion Transformerscited by 158
  24. 2024PuLID: Pure and Lightning ID Customization via Contrastive Alignmentcited by 144
  25. 2024MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidancecited by 144
  26. 2023SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generationcited by 143
  27. 2024FreeCustom: Tuning-Free Customized Image Generation for Multi-Concept Compositioncited by 49
  28. 2024Diffusion Self-Distillation for Zero-Shot Customized Image Generationcited by 47
  29. 2024Resolving Multi-Condition Confusion for Finetuning-Free Personalized Image Generationcited by 43
  30. 2024RealCustom: Narrowing Real Text Word for Real-Time Open-Domain Text-to-Image Customizationcited by 39
  31. 2024BootPIG: Bootstrapping Zero-shot Personalized Image Generation Capabilities in Pretrained Diffusion Modelscited by 27
  32. 2023ViCo: Plug-and-play Visual Condition for Personalized Text-to-image Generationcited by 23
  33. 2024Concept Weaver: Enabling Multi-Concept Fusion in Text-to-Image Modelscited by 23
  34. 2024VMix: Improving Text-to-Image Diffusion Model with Cross-Attention Mixing Controlcited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 79; 2026: 75.

154 papers.

Sort:

By area:

  1. 2025OmniGen2: Towards Instruction-Aligned Multimodal Generationcited by 253
  2. 2025In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformercited by 204
  3. 2025Emu3.5: Native Multimodal Models are World Learnerscited by 132
  4. 2025Echo-4o: Harnessing the Power of GPT-4o Synthetic Images for Improved Image Generationcited by 100
  5. 2025Ovis-U1 Technical Reportcited by 67
  6. 2025XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulationcited by 56
  7. 2025VisualCloze: A Universal Image Generation Framework via Visual In-Context Learningcited by 39
  8. 2025DreamOmni2: Multimodal Instruction-based Editing and Generationcited by 38
  9. 2025CreatiDesign: A Unified Multi-Conditional Diffusion Transformer for Creative Graphic Designcited by 36
  10. 2025USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learningcited by 27
  11. 2025InstantCharacter: Personalize Any Characters with a Scalable Diffusion Transformer Frameworkcited by 19
  12. 2025UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Rewardcited by 19
38 more
  1. 2025FullDiT2: Efficient In-Context Conditioning for Video Diffusion Transformerscited by 18
  2. 2025ContextGen: Contextual Layout Anchoring for Identity-Consistent Multi-Instance Generationcited by 18
  3. 2025MMIG-Bench: Towards Comprehensive and Explainable Evaluation of Multi-Modal Image Generation Modelscited by 17
  4. 2025VINCIE: Unlocking In-context Image Editing from Videocited by 17
  5. 2025WithAnyone: Towards Controllable and ID Consistent Image Generationcited by 17
  6. 2025WordCon: Word-level Typography Control in Scene Text Renderingcited by 16
  7. 2025Cut2Next: Generating Next Shot via In-Context Tuningcited by 15
  8. 2025Query-Kontext: An Unified Multimodal Model for Image Generation and Editingcited by 15
  9. 2025MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglementcited by 14
  10. 2026NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generationcited by 14
  11. 2025Mod-Adapter: Tuning-Free and Versatile Multi-concept Personalization via Modulation Adaptercited by 13
  12. 2025LoRAShop: Training-Free Multi-Concept Image Generation and Editing with Rectified Flow Transformerscited by 12
  13. 2026Lance: Unified Multimodal Modeling by Multi-Task Synergycited by 12
  14. 2025Image Editing As Programs with Diffusion Modelscited by 10
  15. 2025IC-Custom: Diverse Image Customization via In-Context Learningcited by 10
  16. 2025MICo-150K: A Comprehensive Dataset Advancing Multi-Image Compositioncited by 10
  17. 2025Scone: Bridging Composition and Distinction in Subject-Driven Image Generation via Unified Understanding-Generation Modelingcited by 10
  18. 2025Visual-Aware CoT: Achieving High-Fidelity Visual Consistency in Unified Modelscited by 10
  19. 2025MultiBanana: A Challenging Benchmark for Multi-Reference Text-to-Image Generationcited by 9
  20. 2025PSR: Scaling Multi-Subject Personalized Image Generation with Pairwise Subject-Consistency Rewardscited by 9
  21. 2025Image Editing with Diffusion Models: A Surveycited by 8
  22. 2025PairEdit: Learning Semantic Variations for Exemplar-based Image Editingcited by 8
  23. 2025EditIDv2: Editable ID Customization with Data-Lubricated ID Feature Integration for Text-to-Image Generationcited by 8
  24. 2026Re-Align: Structured Reasoning-guided Alignment for In-Context Image Generation and Editingcited by 8
  25. 2025RepText: Rendering Visual Text via Replicatingcited by 7
  26. 2025LAMIC: Layout-Aware Multi-Image Composition via Scalability of Multimodal Diffusion Transformercited by 7
  27. 2025ComposeMe: Attribute-Specific Image Prompts for Controllable Human Image Generationcited by 7
  28. 2026WeEdit: A Dataset, Benchmark and Glyph-Guided Framework for Text-centric Image Editingcited by 7
  29. 2026ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Taskscited by 7
  30. 2025Dimension-Reduction Attack! Video Generative Models are Experts on Controllable Image Synthesiscited by 6
  31. 2025iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generationcited by 6
  32. 2026TAG-MoE: Task-Aware Gating for Unified Generative Mixture-of-Expertscited by 6
  33. 2026InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generationcited by 6
  34. 2026LumosX: Relate Any Identities with Their Attributes for Personalized Video Generationcited by 6
  35. 2025ViStoryBench: Comprehensive Benchmark Suite for Story Visualizationcited by 5
  36. 2025FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalizationcited by 5
  37. 2025SIGMA-GEN: Structure and Identity Guided Multi-subject Assembly for Image Generationcited by 5
  38. 2025ConsistCompose: Unified Multimodal Layout Control for Image Compositioncited by 5

50 most cited of 154 All 154 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Proxy-Tuning: Tailoring Multimodal Autoregressive Models for Subject-Driven Image Generation
  2. 2023FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention
  3. 2026In-Loop Model Adaptation with Coupled Latent-Noise Guidance for High-Fidelity Subject-Driven Text-to-Image Generation
  4. 2025TIDE: Achieving Balanced Subject-Driven Image Generation via Target-Instructed Diffusion Enhancement
  5. 2024Improving Multi-Subject Consistency in Open-Domain Image Generation with Isolation and Reposition Attention
  6. 2025StorySync: Training-Free Subject Consistency in Text-to-Image Generation via Region Harmonization
  7. 2024Obtaining Favorable Layouts for Multiple Object Generation
  8. 2025Context-Aware Autoregressive Models for Multi-Conditional Image Generation