提出几何感知的扩散引导方法，解决高引导强度下的图像失真问题。
Manifold-Optimal Guidance: A Unified Riemannian Control View of Diffusion Guidance
- 将引导建模为流形上的最优控制问题，实现几何自适应更新
- 在不增加计算开销下显著提升图像保真度与结构一致性
- 自动调节引导强度，无需手动调参，适合生成质量敏感场景
无分类器引导（CFG）是条件扩散模型的默认控制机制，但高引导强度常导致过饱和、纹理伪影和结构坍塌。我们归因于几何失配：标准CFG在环境空间中进行欧氏外推，无意间使采样轨迹偏离高密度数据流形。为此，我们提出流形最优引导（MOG），将引导重新建模为局部最优控制问题。MOG提供闭式解的几何感知黎曼更新，无需重训练即可纠正离流形漂移。基于此视角，我们进一步提出Auto-MOG，一种动态能量平衡调度，自适应校准引导强度，有效消除人工超参数调优需求。大量验证表明，MOG相比基线在保真度和对齐性上均有提升，且几乎无额外计算开销。
原文摘要 · Abstract (English)
Classifier-Free Guidance (CFG) serves as the de facto control mechanism for conditional diffusion, yet high guidance scales notoriously induce oversaturation, texture artifacts, and structural collapse. We attribute this failure to a geometric mismatch: standard CFG performs Euclidean extrapolation in ambient space, inadvertently driving sampling trajectories off the high-density data manifold. To resolve this, we present Manifold-Optimal Guidance (MOG), a framework that reformulates guidance as a local optimal control problem. MOG yields a closed-form, geometry-aware Riemannian update that corrects off-manifold drift without requiring retraining. Leveraging this perspective, we further introduce Auto-MOG, a dynamic energy-balancing schedule that adaptively calibrates guidance strength, effectively eliminating the need for manual hyperparameter tuning. Extensive validation demonstrates that MOG yields superior fidelity and alignment compared to baselines, with virtually no added computational overhead.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。