用扩散模型的协方差特征检测分布外数据，效果优于现有方法。
EigenScore: OOD Detection using Covariance in Diffusion Models
- 通过分析扩散模型后验协方差的特征值谱来判断数据是否分布外。
- 在CIFAR-10 vs CIFAR-100等近分布外场景下，性能提升最高达5% AUROC。
- 无需反向传播，仅用前向计算即可高效估计关键特征值，适合实际部署。
分布外（OOD）检测对安全敏感领域中机器学习系统的可靠部署至关重要。扩散模型作为强大的生成模型，可通过迭代去噪捕捉复杂数据分布。受此启发，已有研究探索其在OOD检测中的潜力。本文提出EigenScore，一种利用扩散模型诱导的后验协方差特征值谱进行OOD检测的新方法。我们论证后验协方差能提供一致的分布偏移信号：在分布外输入上，迹和主特征值更大，形成清晰的谱特征。进一步分析明确建立了后验协方差与分布不匹配的关系，证明其作为可靠检测信号的有效性。为保证可计算性，采用无雅可比子空间迭代法，仅通过去噪器前向评估即可估计主导特征值。实验表明，EigenScore达到当前最优性能，相比最佳基线最高提升5% AUROC。尤其在近分布外场景（如CIFAR-10 vs CIFAR-100）中保持鲁棒，而现有扩散基方法常在此类场景失效。
原文摘要 · Abstract (English)
Out-of-distribution (OOD) detection is critical for the safe deployment of machine learning systems in safety-sensitive domains. Diffusion models have recently emerged as powerful generative models, capable of capturing complex data distributions through iterative denoising. Building on this progress, recent work has explored their potential for OOD detection. We propose EigenScore, a new OOD detection method that leverages the eigenvalue spectrum of the posterior covariance induced by a diffusion model. We argue that posterior covariance provides a consistent signal of distribution shift, leading to larger trace and leading eigenvalues on OOD inputs, yielding a clear spectral signature. We further provide analysis explicitly linking posterior covariance to distribution mismatch, establishing it as a reliable signal for OOD detection. To ensure tractability, we adopt a Jacobian-free subspace iteration method to estimate the leading eigenvalues using only forward evaluations of the denoiser. Empirically, EigenScore achieves SOTA performance, with up to 5% AUROC improvement over the best baseline. Notably, it remains robust in near-OOD settings such as CIFAR-10 vs CIFAR-100, where existing diffusion-based methods often fail.
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。