mlsanityData to 14 Oct 2026

ReasonGen-R1: CoT for Autoregressive Image generation models through SFT and RL

arXiv 2505.24875 ↗First posted 30 May 2025cs.CV cs.CLcited by 30

On the map: LLM reasoning → Multimodal reasoning

Abstract (arXiv)

Although chain-of-thought reasoning and reinforcement learning (RL) have driven breakthroughs in NLP, their integration into generative vision models remains underexplored. We introduce ReasonGen-R1, a two-stage framework that first imbues an autoregressive image generator with explicit text-based "thinking" skills via supervised fine-tuning on a newly generated reasoning dataset of written rationales, and then refines its outputs using Group Relative Policy Optimization. To enable the model to reason through text before generating images, We automatically generate and release a corpus of model crafted rationales paired with visual prompts, enabling controlled planning of object layouts, styles, and scene compositions. Our GRPO algorithm uses reward signals from a pretrained vision language model to assess overall visual quality, optimizing the policy in each update. Evaluations on GenEval, DPG, and the T2I benchmark demonstrate that ReasonGen-R1 consistently outperforms strong baselines and prior state-of-the-art models. More: aka.ms/reasongen.

Key facts. Cited by 30 papers in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 18 of its 49 references that are in the map. First posted on arXiv 30 May 2025.

Builds on

18 of 49 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles21 Mar
  2. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization13 Mar
  3. 2025MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning10 Mar
  4. 2025Can We Generate Images with CoT? Let's Verify and Reinforce Image Generation Step by Step23 Jan

Other references · 14

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  4. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
  5. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  6. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  7. 2023PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesiscited by 864
  8. 2024Show-o: One Single Transformer to Unify Multimodal Understanding and Generationcited by 752
  9. 2024Mini-Gemini: Mining the Potential of Multi-modality Vision Language Modelscited by 365
  10. 2025R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimizationcited by 301
  11. 2023DreamLLM: Synergistic Multimodal Comprehension and Creationcited by 297
  12. 2024MetaMorph: Multimodal Understanding and Generation via Instruction Tuningcited by 206
2 more
  1. 2024Dual Diffusion for Unified Image Generation and Understandingcited by 76
  2. 2024ILLUME: Illuminating Your LLMs to See, Draw, and Self-Enhancecited by 64

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 14; 2026: 16.

30 papers.

Sort:

By area:

  1. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  2. 2025Visual Autoregressive Modeling for Instruction-Guided Image Editingcited by 22
  3. 2025Uni-MoE-2.0-Omni: Scaling Language-Centric Omnimodal Large Model with Advanced MoE, Training and Datacited by 21
  4. 2025Dynamic-TreeRPO: Breaking the Independent Trajectory Bottleneck with Structured Samplingcited by 18
  5. 2025Group Critical-token Policy Optimization for Autoregressive Image Generationcited by 17
  6. 2025DraCo: Draft as CoT for Text-to-Image Preview and Rare Concept Generationcited by 13
  7. 2025Integrating Reinforcement Learning with Visual Generative Models: Foundations and Advancescited by 12
  8. 2025Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generationcited by 11
  9. 2025Visual-Aware CoT: Achieving High-Fidelity Visual Consistency in Unified Modelscited by 10
  10. 2025Reinforcement Learning for Large Model: A Surveycited by 9
  11. 2025MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoningcited by 8
  12. 2025Improving Chain-of-Thought Efficiency for Autoregressive Image Generationcited by 7
18 more
  1. 2026From Broad Exploration to Stable Synthesis: Entropy-Guided Optimization for Autoregressive Image Generationcited by 7
  2. 2026Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoningcited by 7
  3. 2025Beyond Textual CoT: Interleaved Text-Image Chains with Deep Confidence Reasoning for Image Editingcited by 6
  4. 2025Expand and Prune: Maximizing Trajectory Diversity for Effective GRPO in Generative Modelscited by 6
  5. 2026Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognitioncited by 6
  6. 2026CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generationcited by 6
  7. 2026Understanding vs. Generation: Navigating Optimization Dilemma in Multimodal Modelscited by 5
  8. 2026Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modelingcited by 5
  9. 2026MiLDEdit: Reasoning-Based Multi-Layer Design Document Editingcited by 3
  10. 2026StruVis: Enhancing Reasoning-based Text-to-Image Generation via Thinking with Structured Visioncited by 3
  11. 2026FiRe: Fine-grained Multimodal Reasoning for Enhanced Image Generationcited by 3
  12. 2026Probing Visual Planning in Image Editing Modelscited by 2
  13. 2026Compositional Text-to-Image Generation Via Region-aware Bimodal Direct Preference Optimizationcited by 2
  14. 2026IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generationcited by 2
  15. 2026GenPuzzle: Benchmarking Visual Reasoning in Image Generation Modelscited by 1
  16. 2026Agentic Visual Generation: From Generative Models to Agentic Controlcited by 1
  17. 2026Uni-Synergy: Bridging Understanding and Generation for Personalized Reasoning via Co-operative Reinforcement Learningcited by 0
  18. 2026Learning Native Reflection in Unified Models with Interleaved Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Delving into RL for Image Generation with CoT: A Study on DPO vs. GRPO
  2. 2025GoT-R1: Unleashing Reasoning Capability of MLLM for Visual Generation with Reinforcement Learning
  3. 2025T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT
  4. 2026UniGRPO: Unified Policy Optimization for Reasoning-Driven Visual Generation
  5. 2025MindOmni: Unleashing Reasoning Generation in Vision Language Models with RGPO
  6. 2026RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection
  7. 2025AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learning
  8. 2026Benchmarking and Evolving Reason-Reflect-Rectify for Reflective Visual Generation