# Towards Unified Multimodal Interleaved Generation via Group Relative Policy Optimization

arXiv: https://arxiv.org/abs/2603.09538 · HTML: https://mlsanity.com/paper/2603.09538/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 34 of its 43 references that are in the map.
- First posted on arXiv 10 Mar 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)

## Builds on (most cited first, 30 of 34)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment](https://mlsanity.com/paper/2310.11513/) (2023)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](https://mlsanity.com/paper/2501.17811/) (2025)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model](https://mlsanity.com/paper/2504.07615/) (2025)
- [Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model](https://mlsanity.com/paper/2408.11039/) (2024)
- [Generative Multimodal Models are In-Context Learners](https://mlsanity.com/paper/2312.13286/) (2023)
- [Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models](https://mlsanity.com/paper/2503.09567/) (2025)
- [SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generation](https://mlsanity.com/paper/2404.14396/) (2024)
- [VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation](https://mlsanity.com/paper/2409.04429/) (2024)
- [Generating Images with Multimodal Language Models](https://mlsanity.com/paper/2305.17216/) (2023)
- [JourneyDB: A Benchmark for Generative Image Understanding](https://mlsanity.com/paper/2307.00716/) (2023)
- [TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generation](https://mlsanity.com/paper/2412.03069/) (2024)
- [Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey](https://mlsanity.com/paper/2503.12605/) (2025)
- [R1-Zero's "Aha Moment" in Visual Reasoning on a 2B Non-SFT Model](https://mlsanity.com/paper/2503.05132/) (2025)
- [LLaVA-Critic: Learning to Evaluate Multimodal Models](https://mlsanity.com/paper/2410.02712/) (2024)
- [ANOLE: An Open, Autoregressive, Native Large Multimodal Models for Interleaved Image-Text Generation](https://mlsanity.com/paper/2407.06135/) (2024)
- [ILLUME: Illuminating Your LLMs to See, Draw, and Self-Enhance](https://mlsanity.com/paper/2412.06673/) (2024)
- [Skywork R1V2: Multimodal Hybrid Reinforcement Learning for Reasoning](https://mlsanity.com/paper/2504.16656/) (2025)
- [EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions](https://mlsanity.com/paper/2409.18042/) (2024)

## Built on by (most cited first, 4 of 4)

- [AlphaGRPO: Unlocking Self-Reflective Multimodal Generation in UMMs via Decompositional Verifiable Reward](https://mlsanity.com/paper/2605.12495/) (2026)
- [ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation](https://mlsanity.com/paper/2608.04436/) (2026)
- [Bridging Modal Isolation in Interleaved Thinking: Supervising Modality Transitions via Stepwise Reinforcement](https://mlsanity.com/paper/2606.12886/) (2026)
- [STBridge: Shared-Target Alignment for Bridging Understanding and Generation in UMMs](https://mlsanity.com/paper/2607.17140/) (2026)

## Abstract (arXiv)

Unified vision-language models have made significant progress in multimodal understanding and generation, yet they largely fall short in producing multimodal interleaved outputs, which is a crucial capability for tasks like visual storytelling and step-by-step visual reasoning. In this work, we propose a reinforcement learning-based post-training strategy to unlock this capability in existing unified models, without relying on large-scale multimodal interleaved datasets. We begin with a warm-up stage using a hybrid dataset comprising curated interleaved sequences and limited data for multimodal understanding and text-to-image generation, which exposes the model to interleaved generation patterns while preserving its pretrained capabilities. To further refine interleaved generation, we propose a unified policy optimization framework that extends Group Relative Policy Optimization (GRPO) to the multimodal setting. Our approach jointly models text and image generation within a single decoding trajectory and optimizes it with our novel hybrid rewards covering textual relevance, visual-text alignment, and structural fidelity. Additionally, we incorporate process-level rewards to provide step-wise guidance, enhancing training efficiency in complex multimodal tasks. Experiments on MMIE and InterleavedBench demonstrate that our approach significantly enhances the quality and coherence of multimodal interleaved generation.