改进扩散模型反演精度，通过动态调整引导尺度减少噪声误差累积。
POLARIS: Projection-Orthogonal Least Squares for Robust and Adaptive Inversion in Diffusion Models
- 将反演问题从误差补偿转为误差源头控制，动态调整每步引导尺度。
- 仅用一行代码显著提升反演隐变量质量，降低噪声近似误差。
- 适合图像编辑与修复任务，对计算开销极小，易集成到现有系统。
基于扩散模型的反演-去噪范式在图像编辑与修复中表现优异。本文重新审视其机制，揭示了一个被忽视的关键因素：噪声近似误差。该误差源于用前一步的预测近似当前步的噪声，导致反演过程中的误差持续累积。为此，提出投影正交最小二乘法（POLARIS），将反演从误差补偿问题转化为误差源头控制问题。不通过优化嵌入或隐变量来抵消漂移，而是将引导尺度 ω 视为逐步可变，并推导出数学上严谨的公式，在每一步最小化反演误差。实验表明，POLARIS仅需一行代码即可显著提升反演隐变量质量，且几乎无性能开销，有效缓解噪声近似误差，一致提升下游任务精度。
原文摘要 · Abstract (English)
The Inversion-Denoising Paradigm, which is based on diffusion models, excels in diverse image editing and restoration tasks. We revisit its mechanism and reveal a critical, overlooked factor in reconstruction degradation: the approximate noise error. This error stems from approximating the noise at step t with the prediction at step t-1, resulting in severe error accumulation throughout the inversion process. We introduce Projection-Orthogonal Least Squares for Robust and Adaptive Inversion (POLARIS), which reformulates inversion from an error-compensation problem into an error-origin problem. Rather than optimizing embeddings or latent codes to offset accumulated drift, POLARIS treats the guidance scale ω as a step-wise variable and derives a mathematically grounded formula to minimize inversion error at each step. Remarkably, POLARIS improves inversion latent quality with just one line of code. With negligible performance overhead, it substantially mitigates noise approximation errors and consistently improves the accuracy of downstream tasks.
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。