aiqbench analysis
This research advances training-free anomaly detection by repurposing the global semantic and attention capabilities of Vision Transformer [CLS] tokens, eliminating the need for manual tuning or retraining. The proposed framework achieves state-of-the-art performance across multiple datasets while offering a plug-and-play solution that integrates seamlessly with existing foundation models. By dynamically weighting local features based on global context, it addresses a key limitation in current approaches that rely solely on independent patch analysis.
DuoAD: Leveraging [CLS] Dual Characteristics for Training-Free Few-Shot Anomaly Detection
arXiv:2607.23924v1 Announce Type: cross Abstract: Vision foundation models have enabled strong training-free anomaly detection (AD). However, most existing approaches rely primarily on independent local patch features, leaving the global contextual information encoded by Vision Transformers (ViTs) underexploited. In this work, we identify the dual characteristics of the ViT [CLS] token: its embedding provides anomaly-invariant global semantic representation, while its attention maps implicitly highlight spatially abnormal regions. Building on this observation, we propose a fully automated AD framework leveraging global context to remove manual tunings. Our framework introduces (1) an automatic augmentation selection strategy driven by [CLS]-level semantic consistency, and (2) an attention-guided feature reweighting mechanism that dynamically adjusts patch contributions according to [CLS] attention saliency. By integrating these components over multi-level features, our method achieves stable anomaly scoring and precise localization without training or parameter tuning. Under the one-shot setting, it achieves Image-AUC scores of 97.7%, 93.2%, and 84.5% on MVTec-AD, VisA, and Real-IAD. Using a single fixed configuration across categories, backbones, and datasets, the method establishes a new state-of-the-art for plug-and-play, training-free anomaly detection while maintaining strong robustness and practical scalability.
Abstract:Vision foundation models have enabled strong training-free anomaly detection (AD). However, most existing approaches rely primarily on independent local patch features, leaving the global contextual information encoded by Vision Transformers (ViTs) underexploited. In this work, we identify the dual characteristics of the ViT [CLS] token: its embedding provides anomaly-invariant global semantic representation, while its attention maps implicitly highlight spatially abnormal regions. Building on this observation, we propose a fully automated AD framework leveraging global context to remove manual tunings. Our framework introduces (1) an automatic augmentation selection strategy driven by [CLS]-level semantic consistency, and (2) an attention-guided feature reweighting mechanism that dynamically adjusts patch contributions according to [CLS] attention saliency. By integrating these components over multi-level features, our method achieves stable anomaly scoring and precise localization without training or parameter tuning. Under the one-shot setting, it achieves Image-AUC scores of 97.7%, 93.2%, and 84.5% on MVTec-AD, VisA, and Real-IAD. Using a single fixed configuration across categories, backbones, and datasets, the method establishes a new state-of-the-art for plug-and-play, training-free anomaly detection while maintaining strong robustness and practical scalability.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2607.23924 [cs.CV]
(or arXiv:2607.23924v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2607.23924 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 277 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
Presentation: Designing AI Platforms for Reliability: Tools for Certainty, Agents for Discovery
InfoQ · May 27, 2026
SEDualVLN: A Spatially-Enhanced Dual-System for Vision-Language Navigation
arXiv · Jun 5, 2026
Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding
arXiv · Jul 29, 2026