mlsanityData to 7 Oct 2026

Multimodal Large Language Models for Multi-Subject In-Context Image Generation

arXiv 2604.07422 ↗First posted 8 Apr 2026cs.LGcited by 2

On the map: Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning

Abstract (arXiv)

Recent advances in text-to-image (T2I) generation have enabled visually coherent image synthesis from descriptions, but generating images containing multiple given subjects remains challenging. As the number of reference identities increases, existing methods often suffer from subject missing and semantic drift. To address this problem, we propose MUSIC, the first MLLM specifically designed for MUlti-Subject In-Context image generation. To overcome the data scarcity, we introduce an automatic and scalable data generation pipeline that eliminates the need for manual annotation. Furthermore, we enhance the model's understanding of multi-subject semantic relationships through a vision chain-of-thought (CoT) mechanism, guiding step-by-step reasoning from subject images to semantics and generation. To mitigate identity entanglement and manage visual complexity, we develop a novel semantics-driven spatial layout planning method and demonstrate its test-time scalability. By incorporating complex subject images during training, we improve the model's capacity for chained reasoning. In addition, we curate MSIC, a new benchmark tailored for multi-subject in-context generation. Experimental results demonstrate that MUSIC significantly surpasses other methods in both multi- and single-subject scenarios.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 51 of its 77 references that are in the map. First posted on arXiv 8 Apr 2026.

Builds on

51 of 77 references.

References · 51

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  4. 2020Denoising Diffusion Implicit Modelscited by 9,655
  5. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  7. 2022Generative Adversarial Networkscited by 5,603
  8. 2017Neural Discrete Representation Learningcited by 4,650
  9. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  10. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  11. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  12. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
38 more
  1. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  2. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  3. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
  4. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  5. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
  6. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  7. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,184
  8. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,122
  9. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,005
  10. 2022In-context Learning and Induction Headscited by 930
  11. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 915
  12. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  13. 2016Pixel Recurrent Neural Networkscited by 814
  14. 2024Autoregressive Model Beats Diffusion: Llama for Scalable Image Generationcited by 760
  15. 2021Multimodal Few-Shot Learning with Frozen Language Modelscited by 657
  16. 2022All are Worth Words: A ViT Backbone for Diffusion Modelscited by 498
  17. 2023BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editingcited by 438
  18. 2024OminiControl: Minimal and Universal Control for Diffusion Transformercited by 334
  19. 2024SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generationcited by 328
  20. 2024Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesiscited by 297
  21. 2023MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizercited by 260
  22. 2022Re-Imagen: Retrieval-Augmented Text-to-Image Generatorcited by 208
  23. 2023Subject-Diffusion:Open Domain Personalized Text-to-Image Generation without Test-time Fine-tuningcited by 191
  24. 2023Making LLaMA SEE and Draw with SEED Tokenizercited by 184
  25. 2025Less-to-More Generalization: Unlocking More Controllability by In-Context Generationcited by 153
  26. 2023What Makes Good Examples for Visual In-Context Learning?cited by 150
  27. 2024Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokenscited by 147
  28. 2023SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generationcited by 142
  29. 2024MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidancecited by 142
  30. 2024MaskBit: Embedding-free Image Generation via Bit Tokenscited by 96
  31. 2024RandAR: Decoder-only Autoregressive Visual Generation in Random Orderscited by 82
  32. 2024Text4Seg: Reimagining Image Segmentation as Text Generationcited by 59
  33. 2024Resolving Multi-Condition Confusion for Finetuning-Free Personalized Image Generationcited by 43
  34. 2024BootPIG: Bootstrapping Zero-shot Personalized Image Generation Capabilities in Pretrained Diffusion Modelscited by 27
  35. 2026Unify-Agent: A Unified Multimodal Agent for World-Grounded Image Synthesiscited by 16
  36. 2024Towards Stabilized and Efficient Diffusion Transformers through Long-Skip-Connections with Spectral Constraintscited by 11
  37. 2025DC-ControlNet: Decoupling Inter- and Intra-Element Conditions in Image Generation with Diffusion Modelscited by 10
  38. 2026Condition Errors Refinement in Autoregressive Image Generation with Diffusion Losscited by 9

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026CogOmniControl: Reasoning-Driven Controllable Video Generation via Creative Intent Cognitioncited by 0
  2. 2026Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025MM-R1: Unleashing the Power of Unified Multimodal Large Language Models for Personalized Image Generation
  2. 2026Unified Text-Image Generation with Weakness-Targeted Post-Training
  3. 2025CINEMA: Coherent Multi-Subject Video Generation via MLLM-Based Guidance
  4. 2025MUSE: Multi-Subject Unified Synthesis via Explicit Layout Semantic Expansion
  5. 2025CRAFT: Continuous Reasoning and Agentic Feedback Tuning for Multimodal Text-to-Image Generation
  6. 2025GoT: Unleashing Reasoning Capability of Multimodal Large Language Model for Visual Generation and Editing
  7. 2023Kosmos-G: Generating Images in Context with Multimodal Large Language Models
  8. 2024Can MLLMs Perform Text-to-Image In-Context Learning?