面向跨域异常检测的异常相关残差场
Anomaly-Related Residual Fields for Cross-domain Anomaly Detection
把扩散模型反向过程的残差按时间步演化建模为「残差演化场」，用平稳性区分正常变异与异常信号，这一思路可直接迁移到任何基于重建/生成残差的视觉异常检测与OOD检测流程中，并配合跨域场对齐实现无标签目标域适配。
原文依据 anomalous regions retain an additional non-stationary signal that persists
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论无标签图像异常检测之所以困难，是因为必须把异常与正常类内的变异区分开来。扩散模型为正常数据学习了一个流形，并且在「偏离流形的异常更难被生成、因而产生更大预测误差」这一常见假设下，许多方法基于预测残差构建检测器；然而反向过程的随机性以及复杂但正常的结构同样会产生大残差，因此仅凭幅值大小并不具备诊断性。为厘清从这种带噪残差中究竟能恢复出什么，本文从理论上考察残差信号如何在后续反向步骤中传播，结果表明：与正常统计相一致的变异会逐渐被吸收而趋于平稳，而异常区域则保留一个持续存在的额外非平稳信号。基于这一洞见，残差演化场（REF）将该持续信号分离出来，用带标签的源数据校准提取器，并通过跨域场对齐（CFA）将其迁移到无标签的目标域。本文建立了包含形式化分析的理论框架，在显著域偏移下的多个基准上实验表明达到最优性能，相较强基线提升 2.01 至 14 个百分点（pp）。
英文原文摘要 1267 字符
Label-free image anomaly detection is difficult because anomalies must be separated from intra-normal variability. Diffusion models learn a manifold for normal data, and, under the common assumption that off-manifold anomalies are harder to generate and yield larger prediction errors, many methods build detectors from prediction residuals; yet reverse-process stochasticity and complex but normal structure also produce large residuals, so magnitude alone is non-diagnostic. To clarify what is recoverable from such noisy residuals, the theory examines how residual signals propagate through later reverse steps, showing that variability consistent with normal statistics is gradually absorbed toward stationarity, whereas anomalous regions retain an additional non-stationary signal that persists. Building on this insight, the Residual-Evolution Field (REF) isolates this persistent signal, with labeled source data calibrating the extractor and Cross-domain Field Alignment (CFA) transferring it to unlabeled targets. A theoretical framework with formal analysis is established, and experiments across multiple benchmarks under substantial domain shifts demonstrate state-of-the-art performance, improving over strong baselines by 2.01-14 percentage points (pp).