AI画图终于能边写边画了
现在的AI画图工具（如Midjourney、FLUX）能生成单张惊艳的照片，但你要它“先画一只猫，再写一段描述，再画猫在沙发上”这种图文交替的序列，它就卡住了。这篇论文搞了个多智能体流水线：一个“规划师”把图文顺序拆成步骤，指挥画图模型一步步执行；一个“评论家”检查每一步的结果，如果画歪了就修正指令重画。在图文交替生成测试中，效果追平了GPT-5。更意外的是，这个流程还让基础模型在推理类任务上变强了——比如4步FLUX模型在WISE和RISE测试中大幅提升。它不是你能明天装上的插件，但指明了让AI真正“边想边画”的方向。
📄 原文摘要(英文)
Recent image generators have demonstrated impressive photorealism and instruction-following capabilities in single-image generation and editing. However, constrained by their architectures, they cannot achieve interleaved generation (text-image sequence), which has crucial applications in visual narratives, guidance, and embodied manipulation. Even the latest open-source Unified Multimodal Models (UMMs) exhibit limited performance in this regard. In this paper, we introduce InterleaveThinker, the first multi-agent pipeline designed to endow any existing image generator with interleaved generation capabilities. Specifically, we employ a planner agent to organize the image-text input sequence, instructing the image generator on the required execution at each step. Subsequently, we introduce a critic agent to evaluate the generator's outputs, identify samples that deviate from the planned instructions, and refine the instructions for regeneration. To implement this pipeline, we construct the Interleave-Planner-SFT-80k and Interleave-Critic-SFT-112k to perform a format cold-start. Then we develop Interleave-Critic-RL-13k to reinforce the step-wise instruction correction capability within a generation trajectory using GRPO. Since a single interleaved generation trajectory may involve over 25 generator calls, optimizing the entire trajectory is computationally impractical. Therefore, we propose accuracy reward and step-wise reward, allowing single-step RL to effectively guide the entire generation trajectory. The results show that InterleaveThinker improves performance across various image generators. On interleaved generation benchmarks, it achieves performance comparable to Nano Banana and GPT-5. Surprisingly, it also significantly enhances the base model on reasoning-based benchmarks; for example, on 4-step FLUX.2-klein, we observe substantial gains on WISE and RISE.