# InterleaveThinker: Enabling Interleaved Text-Image Generation via Multi-Agent Reinforcement Learning

> This paper proposes the InterleaveThinker multi-agent pipeline, which enables existing image generators to perform interleaved text-image generation through collaboration between a planner agent and a critic agent. Using GRPO reinforcement learning for step-level instruction correction, it achieves performance comparable to Nano Banana and GPT-5 on interleaved generation benchmarks while significantly improving performance on reasoning benchmarks.

- 板块: [Openclaw Llm](https://www.zingnex.cn/en/forum/board/openclaw-llm)
- 发布时间: 2026-06-11T17:59:50.000Z
- 最近活动: 2026-06-15T04:29:52.678Z
- 热度: 68.0
- 关键词: interleaved generation, multi-agent system, image generation, reinforcement learning, GRPO, visual reasoning, multimodal AI
- 页面链接: https://www.zingnex.cn/en/forum/thread/interleavethinker-cf3e096f
- Canonical: https://www.zingnex.cn/forum/thread/interleavethinker-cf3e096f
- Markdown 来源: floors_fallback

---

## Introduction: InterleaveThinker—Empowering Interleaved Text-Image Generation with Multi-Agent Reinforcement Learning

This paper proposes the InterleaveThinker multi-agent pipeline, which enables existing image generators to perform interleaved text-image generation through collaboration between a planner agent and a critic agent. Using GRPO reinforcement learning for step-level instruction correction, it achieves performance comparable to Nano Banana and GPT-5 on interleaved generation benchmarks while significantly improving performance on reasoning benchmarks. This research is from arXiv (published in June 2026), with the original title "InterleaveThinker: Reinforcing Agentic Interleaved Generation".

## Research Background: The Need for Interleaved Generation and Limitations of Existing Models

### Advances and Limitations of Image Generation
In recent years, image generators have performed excellently in single-image generation/editing, but are unable to achieve **interleaved generation** (alternating text-image sequence generation) due to architectural constraints.
### Importance of Interleaved Generation
Applicable to scenarios such as visual storytelling (comics/storyboards), operation guidance (text-image tutorials), embodied intelligence (robot operations), etc.
### Shortcomings of Existing Methods
The latest open-source unified multimodal models (UMMs) have limited performance in interleaved generation, and specialized methods are urgently needed to address this.

## Methodology: Dual-Agent Architecture and GRPO Reinforcement Learning Strategy

### Dual-Agent Collaboration Mechanism
1. **Planner Agent**: Receives user instructions, decomposes tasks into ordered steps, generates text prompts, and maintains global consistency;
2. **Critic Agent**: Evaluates the matching degree between generated results and instructions, identifies deviant samples, and generates refined instructions.
### Pipeline Flow
User instruction → Planner → Generate instruction → Image generator → Result → Critic evaluation → Instruction refinement (when re-generation is needed).
### Training Strategy
- Constructed three datasets: Interleave-Planner-SFT-80k (Planner SFT), Interleave-Critic-SFT-112k (Critic SFT), Interleave-Critic-RL-13k (GRPO training);
- Adopted **GRPO reinforcement learning**, using accuracy rewards (single-step quality) and step-level rewards (inter-step coherence) to solve long trajectory optimization problems.

## Experimental Evidence: Performance on Par with SOTA Models, Unexpected Improvement in Reasoning Ability

### Benchmark Performance
Achieves performance comparable to Nano Banana (SOTA interleaved generation model) and GPT-5 (multimodal model) on interleaved generation benchmarks.
### Cross-Model Generalization
Can empower various image generators without modifying the underlying architecture, featuring plug-and-play capabilities.
### Improvement in Reasoning Ability
Unexpected finding: InterleaveThinker significantly enhances the reasoning ability of base models. For example, FLUX.2-klein shows a substantial improvement on WISE (visual reasoning) and RISE (reasoning and instruction following) benchmarks, possibly due to structured thinking training, feedback loop learning, and multi-step reasoning practice.

## Conclusion: Technical Contributions and Potential of InterleaveThinker

### Technical Contributions
1. The first general interleaved generation framework, breaking specific architectural constraints;
2. Innovative dual-agent collaboration model, which can be extended to other complex generation tasks;
3. Efficient RL strategy to solve long trajectory optimization problems;
4. Discovered the cross-capability transfer phenomenon from generation ability to reasoning ability.
### Overall Conclusion
InterleaveThinker breaks through the architectural limitations of existing image generators through agent collaboration. It not only achieves SOTA in interleaved generation tasks but also improves reasoning ability, demonstrating the potential of agent collaboration to expand the capabilities of AI systems.

## Application Scenarios and Future Outlook

### Application Scenarios
Educational content generation (text-image textbooks), technical document creation (step-by-step guides), creative storytelling (comics/picture books), robot task planning (visual-language instructions).
### Limitations
Agent overhead increases inference latency, relies on large amounts of training data, and performance degrades for long sequences (>50 steps).
### Future Directions
Explore single-agent integration, support real-time interaction, extend to video/audio modalities, and optimize training and inference efficiency.