mlsanityData to 14 Oct 2026

ImageGen-CoT: Enhancing Text-to-Image In-context Learning with Chain-of-Thought Reasoning

arXiv 2503.19312 ↗First posted 25 Mar 2025cs.CVcited by 37

On the map: LLM reasoning → Multimodal reasoning

Abstract (arXiv)

In this work, we study the problem of Text-to-Image In-Context Learning (T2I-ICL). While Unified Multimodal LLMs (MLLMs) have advanced rapidly in recent years, they struggle with contextual reasoning in T2I-ICL scenarios. To address this limitation, we propose a novel framework that incorporates a thought process called ImageGen-CoT prior to image generation. To avoid generating unstructured ineffective reasoning steps, we develop an automatic pipeline to curate a high-quality ImageGen-CoT dataset. We then fine-tune MLLMs using this dataset to enhance their contextual reasoning capabilities. To further enhance performance, we explore test-time scale-up strategies and propose a novel hybrid scaling approach. This approach first generates multiple ImageGen-CoT chains and then produces multiple images for each chain via sampling. Extensive experiments demonstrate the effectiveness of our proposed method. Notably, fine-tuning with the ImageGen-CoT dataset leads to a substantial 80% performance gain for SEED-X on T2I-ICL tasks. See our project page at https://ImageGen-CoT.github.io/. Code and model weights will be open-sourced.

Key facts. Cited by 37 papers in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 38 of its 45 references that are in the map. First posted on arXiv 25 Mar 2025.

Builds on

38 of 45 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization15 Nov

Other references · 37

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  4. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  5. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  6. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  7. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  8. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  9. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
  10. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  11. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  12. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
25 more
  1. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,245
  2. 2024Qwen2 Technical Reportcited by 2,120
  3. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  4. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,106
  5. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  6. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 927
  7. 2024Emu3: Next-Token Prediction is All You Needcited by 781
  8. 2024Show-o: One Single Transformer to Unify Multimodal Understanding and Generationcited by 752
  9. 2024DeepSeek-VL: Towards Real-World Vision-Language Understandingcited by 733
  10. 2023The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)cited by 681
  11. 2021Multimodal Few-Shot Learning with Frozen Language Modelscited by 661
  12. 2022GIT: A Generative Image-to-text Transformer for Vision and Languagecited by 537
  13. 2023Generative Multimodal Models are In-Context Learnerscited by 446
  14. 2024SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generationcited by 331
  15. 2023Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Actioncited by 302
  16. 2021An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQAcited by 301
  17. 2023DreamLLM: Synergistic Multimodal Comprehension and Creationcited by 297
  18. 2023Multimodal Foundation Models: From Specialists to General-Purpose Assistantscited by 247
  19. 2024MetaMorph: Multimodal Understanding and Generation via Instruction Tuningcited by 206
  20. 2024TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generationcited by 202
  21. 2023Making LLaMA SEE and Draw with SEED Tokenizercited by 185
  22. 2024DreamBench++: A Human-Aligned Benchmark for Personalized Image Generationcited by 131
  23. 2024SynerGen-VL: Towards Synergistic Image Understanding and Generation with Vision Experts and Token Foldingcited by 42
  24. 2024Can MLLMs Perform Text-to-Image In-Context Learning?cited by 13
  25. 2024CAT: Contrastive Adapter Training for Personalized Image Generationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 19; 2026: 18.

37 papers.

Sort:

By area:

  1. 2025VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Modelscited by 98
  2. 2025Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?cited by 34
  3. 2025RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmarkcited by 33
  4. 2025UniGen: Enhanced Training & Test-Time Strategies for Unified Multimodal Understanding and Generationcited by 32
  5. 2025R2I-Bench: Benchmarking Reasoning-Driven Text-to-Image Generationcited by 32
  6. 2026GEMS: Agent-Native Multimodal Generation with Memory and Skillscited by 21
  7. 2025Thinking-while-Generating: Interleaving Textual Reasoning throughout Visual Generationcited by 18
  8. 2025MMMG: A Massive, Multidisciplinary, Multi-Tier Generation Benchmark for Text-to-Image Reasoningcited by 16
  9. 2025Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generationcited by 11
  10. 2025SuperFlow: Training Flow Matching Models with RL on the Flycited by 10
  11. 2025Visual Instruction Pretraining for Domain-Specific Foundation Modelscited by 9
  12. 2025ControlThinker: Unveiling Latent Semantics for Controllable Image Generation through Visual Reasoningcited by 8
25 more
  1. 2026Show, Don't Tell: Morphing Latent Reasoning into Image Generationcited by 8
  2. 2025Improving Chain-of-Thought Efficiency for Autoregressive Image Generationcited by 7
  3. 2026From Broad Exploration to Stable Synthesis: Entropy-Guided Optimization for Autoregressive Image Generationcited by 7
  4. 2026Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoningcited by 7
  5. 2025Beyond Textual CoT: Interleaved Text-Image Chains with Deep Confidence Reasoning for Image Editingcited by 6
  6. 2025Visual-CoG: Stage-Aware Reinforcement Learning with Chain of Guidance for Text-to-Image Generationcited by 5
  7. 2026Unified Thinker: A General Reasoning Modular Core for Image Generationcited by 5
  8. 2026SPIRAL: Self-Evolving Action-Conditioned Video Generation via Reflective Planning Agentscited by 5
  9. 2025POET: Preference Optimization for Enhanced Text-to-Image Generationcited by 4
  10. 2025EmoCtrl: Controllable Emotional Image Content Generationcited by 4
  11. 2026PhysAlign: Physics-Coherent Image-to-Video Generation through Feature and 3D Representation Alignmentcited by 4
  12. 2026MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generationcited by 4
  13. 2026StruVis: Enhancing Reasoning-based Text-to-Image Generation via Thinking with Structured Visioncited by 3
  14. 2026DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editingcited by 3
  15. 2026Breaking Dual Bottlenecks: Evolving Unified Multimodal Models into Self-Adaptive Interleaved Visual Reasonerscited by 3
  16. 2025OSPO: Object-Centric Self-Improving Preference Optimization for Text-to-Image Generationcited by 1
  17. 2025Rethinking Prompt Design for Inference-time Scaling in Text-to-Visual Generationcited by 1
  18. 2026Unlocking Complex Visual Generation via Closed-Loop Verified Reasoningcited by 1
  19. 2026Latent Action Control for Reasoning-Guided Unified Image Generationcited by 1
  20. 2026SciIR: A Large-scale Training Dataset and Benchmark for Scientific Image Reasoning Generationcited by 1
  21. 2025BatonVoice: An Operationalist Framework for Enhancing Controllable Speech Synthesis with Linguistic Intelligence from LLMscited by 0
  22. 2026Are Text-to-Image Models Inductivist Turkeys? A Counterfactual Benchmark for Causal Reasoningcited by 0
  23. 2026Tree-of-Thoughts Reasoning for Text-to-Image In-Context Learningcited by 0
  24. 2026DiSCO: Defending text-to-image generation through distribution-guided contrastive prompt optimizationcited by 0
  25. 2026Open-UniMo: Towards Unified Motion-Language Understanding and Generation in the Open Worldcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Unsupervised Visual Chain-of-Thought Reasoning via Preference Optimization
  2. 2026TVI-CoT: Text-Visual Interleaved Chain-of-Thought Reasoning for Multimodal Understanding
  3. 2025T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT
  4. 2025Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Vision
  5. 2025Will Pre-Training Ever End? A First Step Toward Next-Generation Foundation MLLMs via Self-Improving Systematic Cognition
  6. 2026CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generation
  7. 2024Evaluating Linguistic Capabilities of Multimodal LLMs in the Lens of Few-Shot Learning
  8. 2026Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning