通过坐标曲率差定位扩散模型中的记忆区域，提升隐私检测精度。
Localizing Memorized Regions in Diffusion Models via Coordinate-Wise Curvature Differences
- 利用坐标级曲率差异识别生成图像中的记忆区域。
- 在Stable Diffusion上定位准确率超越现有注意力方法。
- 适合关注生成模型隐私与版权问题的研究者使用。
扩散模型可能无意中记忆训练样本，引发隐私和版权担忧。尽管已有方法可检测记忆现象，但多依赖全局或模型特定信号，难以定位记忆在图像中的具体位置。本文从几何角度将局部记忆表征为坐标级方差坍缩，并提出曲率差方法，通过减去欠拟合基线（如无条件模型或训练不足版本）的曲率，分离出过拟合导致的记忆。进一步推导出得分差代理，为广泛使用的得分差检测指标提供几何解释。在Stable Diffusion上的实验表明，该方法在真实记忆掩码下的定位性能优于先前基于注意力的方法。代码已公开于https://github.com/Gwangho99/mem-curv-diff。
原文摘要 · Abstract (English)
Diffusion models can unintentionally memorize training samples, raising concerns about privacy and copyright. While recent methods can detect memorization, they often rely on global or model-specific signals and provide limited insight into where memorization appears within a generated image. We provide a geometric characterization of local memorization as a coordinate-wise variance collapse. However, such collapse can also arise from intrinsic data constraints rather than overfitting. To isolate overfitting-driven memorization, we propose curvature-difference methods that subtract the curvature of an underfitted baseline, either the unconditional model or a less-trained version of itself. We further derive a score-difference proxy that provides a geometric explanation for the widely used score-difference-based detection metric. Experiments on Stable Diffusion, evaluated against ground-truth memorization masks, show that our method outperforms the prior attention-based localization method. Code is available at https://github.com/Gwangho99/mem-curv-diff.
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。