合成课程强化组合式文本到图像生成
Synthetic Curriculum Reinforces Compositional Text-to-Image Generation
用场景图定义组合难度并配合 MCMC 图采样构造由易到难的训练课程，再接入 GRPO 做强化学习微调，这套「难度感知课程 + 策略优化」流程可直接迁移到扩散或自回归 T2I 模型的组合能力提升上。
原文依据 we leverage scene graphs to establish a novel difficulty criterion for compositional ability and develop a corresponding adaptive Markov Chain Monte Carlo graph sampling algorithm
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论文本到图像（T2I）生成长期以来是一个开放问题，其中组合式合成仍然尤其具有挑战性。该任务要求准确渲染包含多个对象、且这些对象具有多样属性以及复杂空间与语义关系的复杂场景，既需要精确的物体放置，也需要连贯的物体间交互。本文提出一种新颖的组合式课程强化学习框架 CompGen，用以解决现有 T2I 模型在组合能力上的不足。具体而言，我们利用场景图建立了一种新的组合能力难度判据，并开发了相应的自适应马尔可夫链蒙特卡洛图采样算法。这种难度感知的方法能够合成训练课程数据，通过强化学习逐步优化 T2I 模型。我们将课程学习方法集成到组相对策略优化（GRPO）中，并研究了不同的课程调度策略。实验表明，CompGen 在不同课程调度策略下呈现出不同的缩放曲线，其中由易到难策略与高斯采样策略相比随机采样取得了更优的缩放性能。大量实验表明，CompGen 显著增强了基于扩散和自回归的 T2I 模型的组合式生成能力，凸显了其在改进组合式 T2I 生成系统方面的有效性。
英文原文摘要 1535 字符
Text-to-Image (T2I) generation has long been an open problem, with compositional synthesis remaining particularly challenging. This task requires accurate rendering of complex scenes containing multiple objects that exhibit diverse attributes as well as intricate spatial and semantic relationships, demanding both precise object placement and coherent inter-object interactions. In this paper, we propose a novel compositional curriculum reinforcement learning framework named CompGen that addresses compositional weakness in existing T2I models. Specifically, we leverage scene graphs to establish a novel difficulty criterion for compositional ability and develop a corresponding adaptive Markov Chain Monte Carlo graph sampling algorithm. This difficulty-aware approach enables the synthesis of training curriculum data that progressively optimize T2I models through reinforcement learning. We integrate our curriculum learning approach into Group Relative Policy Optimization (GRPO) and investigate different curriculum scheduling strategies. Our experiments reveal that CompGen exhibits distinct scaling curves under different curriculum scheduling strategies, with easy-to-hard and Gaussian sampling strategies yielding superior scaling performance compared to random sampling. Extensive experiments demonstrate that CompGen significantly enhances compositional generation capabilities for both diffusion-based and auto-regressive T2I models, highlighting its effectiveness in improving the compositional T2I generation systems.