跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2605.28630 · v2

Relation-Aware Evidence Routing and Confidence-Guided Adaptation for Zero-Shot Visual Anomaly Detection

首次提交 2026-05-27 23:38
版本更新 2026-11-17 22:32
雷达观测 2026-11-18 16:38
arXiv ↗PDF ↗

摘要

Zero-shot visual anomaly detection (ZSAD) aims to identify and localize anomalies in unseen categories without target-domain adaptation, yet heterogeneous anomaly patterns and cross-domain variations challenge reliable detection. Existing approaches primarily rely on local visual features, patch-text similarities, or memory-based retrieval, leaving inter-patch relational information underexplored in anomaly evidence modeling. In this paper, we propose RACE-AD, a relation-aware evidence routing and confidence-guided adaptation framework that integrates local anomaly responses with attention-derived relational information for generalizable ZSAD. To capture inter-patch relationships, we characterize the dispersion of visual attention distributions using Shannon entropy, providing complementary relational statistics beyond local anomaly evidence. To organize heterogeneous evidence, we introduce relation-aware routing that combines memory-derived normal/anomaly responses with attention entropy to construct image-conditioned representations through entropy-regularized soft aggregation. To mitigate unreliable retrieval responses under domain shift, we design confidence-guided adaptation that regulates anomaly-conditioned representations according to the strength of the memory anomaly evidence and the relational dispersion before visual-semantic alignment. Extensive experiments on ten industrial and medical benchmarks demonstrate the effectiveness of RACE-AD, achieving average image-level AUROC and AP scores of 91.9% and 93.7%, and pixel-level AUROC and AUPRO scores of 93.6% and 80.0%, respectively. Ablation and diagnostic analyses show that entropy-guided routing mainly improves region-level localization and that the confidence gate is essential under domain shift.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 22:32 · 最新观测版本

Relation-Aware Evidence Routing and Confidence-Guided Adaptation for Zero-Shot Visual Anomaly Detection

Zero-shot visual anomaly detection (ZSAD) aims to identify and localize anomalies in unseen categories without target-domain adaptation, yet heterogeneous anomaly patterns and cross-domain variations challenge reliable detection. Existing approaches primarily rely on local visual features, patch-text similarities, or memory-based retrieval, leaving inter-patch relational information underexplored in anomaly evidence modeling. In this paper, we propose RACE-AD, a relation-aware evidence routing and confidence-guided adaptation framework that integrates local anomaly responses with attention-derived relational information for generalizable ZSAD. To capture inter-patch relationships, we characterize the dispersion of visual attention distributions using Shannon entropy, providing complementary relational statistics beyond local anomaly evidence. To organize heterogeneous evidence, we introduce relation-aware routing that combines memory-derived normal/anomaly responses with attention entropy to construct image-conditioned representations through entropy-regularized soft aggregation. To mitigate unreliable retrieval responses under domain shift, we design confidence-guided adaptation that regulates anomaly-conditioned representations according to the strength of the memory anomaly evidence and the relational dispersion before visual-semantic alignment. Extensive experiments on ten industrial and medical benchmarks demonstrate the effectiveness of RACE-AD, achieving average image-level AUROC and AP scores of 91.9% and 93.7%, and pixel-level AUROC and AUPRO scores of 93.6% and 80.0%, respectively. Ablation and diagnostic analyses show that entropy-guided routing mainly improves region-level localization and that the confidence gate is essential under domain shift.

查看此版本解读 →查看此版本原文 ↗