发表机构
Fudan University（复旦大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出Anchor and Adapt两阶段提示学习框架，分离异常语义获取与目标正常外观适配，通过固定锚点和适配正常分支，在少样本工业异常检测中实现竞争性能，减少对类别特定模板的依赖。
AI 中文摘要
在少样本工业异常检测中，少量正常目标图像不提供直接的缺陷监督，使得仅从这些样本中学习异常提示变得困难。因此，一些视觉-语言方法使用手动指定的描述来提供明确的异常语义。然而，构建这些描述需要针对特定产品的努力，且其有效性依赖于提示选择。我们提出锚定与适配（Anchor and Adapt），一种两阶段提示学习框架，将异常语义的获取与对目标正常外观的适配分离。第一阶段从标注的辅助数据中学习可迁移的正常和异常锚点。第二阶段保持这些锚点固定，并使用少量目标正常样本适配额外的正常分支。继承的和适配的正常分支共同表征目标正常性，文本锚点正则化鼓励与通用正常先验的一致性以及与异常锚点的分离。该设计保留了学习到的异常知识，同时减少了对类别特定异常模板的依赖，且无需合成异常生成。在MVTec-AD和VisA上，在1、2和4样本设置下的跨数据集实验展示了有竞争力的检测和定位性能。受控消融评估了迁移锚点、非对称适配、双正常表示和锚点正则化的作用。
英文摘要
In few-shot industrial anomaly detection, the few normal target images provide no direct defect supervision, making anomaly prompts difficult to learn from these samples alone. Some vision-language methods therefore use manually specified descriptions to supply explicit anomaly semantics. However, constructing these descriptions requires product-specific effort, and their effectiveness depends on prompt selection. We propose Anchor and Adapt, a two-stage prompt learning framework that separates the acquisition of anomaly semantics from adaptation to target normal appearance. Stage I learns transferable normal and abnormal anchors from annotated auxiliary data. Stage II keeps these anchors fixed and adapts an additional normal branch using the few target normal samples. The inherited and adapted normal branches jointly characterize target normality, with text-anchor regularization encouraging consistency with the generic normal prior and separation from the abnormal anchors. This design retains learned anomaly knowledge while reducing dependence on category-specific anomaly templates, without requiring synthetic anomaly generation. Cross-dataset experiments between MVTec-AD and VisA under 1-, 2-, and 4-shot settings demonstrate competitive detection and localization performance. Controlled ablations assess the roles of transferred anchors, asymmetric adaptation, dual-normal representations, and anchor regularization.
