基于内在反思的生成式空间推理器：RL-RIG
RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection
AI总结：
RL-RIG通过生成-反思-编辑范式，提升图像生成的精细空间关系捕捉能力，采用Scene Graph IoU和VLM评估策略，在空间一致性上优于现有模型。
摘要：
近期图像生成技术在高质量图像生成方面取得了显著进展，但现有模型仍面临空间推理难题，难以准确捕捉提示中的细粒度空间关系并生成结构完整的场景。为缓解此问题，我们提出了RL-RIG，一种基于反思的图像生成强化学习框架。该架构包含四个主要组件：Diffuser、Checker、Actor和Inverse Diffuser，遵循生成-反思-编辑范式，以激发图像生成中的链式思维能力。为增强模型对生成轨迹的直觉，我们进一步开发了Reflection-GRPO，用于训练VLM Actor处理编辑提示和Image Editor提升图像质量。与传统方法仅生成视觉吸引人但结构不合理的内容不同，我们的评估指标优先考虑空间准确性，利用Scene Graph IoU和VLM作为评判者策略，在LAION-SG数据集上评估生成图像的空间一致性。实验结果表明，RL-RIG在图像生成的可控性和精确空间推理方面，相较于现有最先进的开源模型，性能提升了高达11%。
英文摘要：
Recent advancements in image generation have achieved impressive results in producing high-quality images. However, existing image generation models still generally struggle with a spatial reasoning dilemma, lacking the ability to accurately capture fine-grained spatial relationships from the prompt and correctly generate scenes with structural integrity. To mitigate this dilemma, we propose RL-RIG, a Reinforcement Learning framework for Reflection-based Image Generation. Our architecture comprises four primary components: Diffuser, Checker, Actor, and Inverse Diffuser, following a Generate-Reflect-Edit paradigm to spark the Chain of Thought reasoning ability in image generation for addressing the dilemma. To equip the model with better intuition over generation trajectories, we further develop Reflection-GRPO to train the VLM Actor for edit prompts and the Image Editor for better image quality under a given prompt, respectively. Unlike traditional approaches that solely produce visually stunning yet structurally unreasonable content, our evaluation metrics prioritize spatial accuracy, utilizing Scene Graph IoU and employing a VLM-as-a-Judge strategy to assess the spatial consistency of generated images on LAION-SG dataset. Experimental results show that RL-RIG outperforms existing state-of-the-art open-source models by up to 11% in terms of controllable and precise spatial reasoning in image generation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
