DART：面向通用异常检测的领域无关残差迁移
DART: Domain-Agnostic Residual Transfer for Generalist Anomaly Detection
Muhammad Aqeel · Maham Nazir · Marco Cristani · Francesco Setti
中文摘要
通用异常检测（GAD）旨在仅用少量参考样本检测未见领域中的异常。近期基于正常-异常引导的方法通过利用异常参考样本改进了这一设定，但其残差往往将可迁移的异常线索与领域特定的外观因素（如纹理、几何和采集条件）混杂在一起。这种残差领域纠缠限制了跨域迁移，并可能在新领域上产生虚假激活。提出 DART，一种用于正常-异常引导 GAD 的插件式框架，学习将残差分离为异常相关成分和领域特定成分。DART 结合残差解耦模块（在抑制领域信息的同时保留异常判别信息）与跨域元训练（在源领域间构建片段以使分离有效）。该方法与架构无关，可插入现有 GAD 流水线的残差计算与检测头之间。在 MVTec AD、VisA 和 BraTS 上的实验表明，相对多种基础方法取得一致提升，在更强领域偏移下增益最大。
关键要点
- 01问题：现有正常-异常引导 GAD 的残差混杂领域外观因素，限制跨域迁移并产生虚假激活。
- 02方法：DART 插件框架用残差解耦模块分离异常相关与领域特定成分，配合跨域元训练。
- 03结果：在 MVTec AD、VisA、BraTS 上相对多种基础方法一致提升，强领域偏移下增益最大。
- 04局限：摘要未提及具体局限。
解读
尚无解读。
原始英文摘要
arXiv:2611.08271v1 Announce Type: new Abstract: Generalist Anomaly Detection (GAD) aims to detect anomalies in unseen domains using only a few reference samples. Recent normal-abnormal guided methods improve this setting by exploiting abnormal references, but their residuals often mix transferable anomaly cues with domain-specific appearance factors such as texture, geometry, and acquisition conditions. This residual domain entanglement limits cross-domain transfer and may produce spurious activations on novel domains. We propose DART, a plug-in framework for normal-abnormal guided GAD that learns to separate residuals into anomaly-relevant and domain-specific components. DART combines a Residual Disentanglement Module, which preserves anomaly-discriminative information while suppressing domain information, with Cross-Domain Meta-Training, which forms episodes across source domains to make the separation effective. The method is architecture-agnostic and can be inserted between residual computation and the detection head of existing GAD pipelines. Experiments on MVTec AD, VisA, and BraTS show consistent improvements over multiple base methods, with the largest gains under stronger domain shifts. https://github.com/aqeeelmirza/DART