arXiv:2506.09237v2 Announce Type: replace-cross 
Abstract: Anomaly Detection (AD) and Anomaly Localization (AL) are crucial in fields that demand high reliability, such as medical imaging and industrial monitoring. However, current AD and AL approaches are often susceptible to adversarial attacks due to limitations in training data, which typically include only normal, unlabeled samples. This study introduces PatchGuard, an adversarially robust AD and AL method that incorporates pseudo anomalies with localization masks within a Vision Transformer (ViT)-based architecture to address these vulnerabilities. We begin by examining the essential properties of pseudo anomalies, and follow it by providing theoretical insights into the attention mechanisms required to enhance the adversarial robustness of AD and AL systems. We then present our approach, which leverages Foreground-Aware Pseudo-Anomalies to overcome the deficiencies of previous anomaly-aware methods. Our method incorporates these crafted pseudo-anomaly samples into a ViT-based framework, with adversarial training guided by a novel loss function designed to improve model robustness, as supported by our theoretical analysis. Experimental results on well-established industrial and medical datasets demonstrate that PatchGuard significantly outperforms previous methods in adversarial settings, achieving performance gains of $53.2\%$ in AD and $68.5\%$ in AL, while also maintaining competitive accuracy in non-adversarial settings. The code repository is available at https://github.com/rohban-lab/PatchGuard .

تقدم دراسة جديدة PatchGuard، وهي طريقة مبتكرة للكشف عن الشذوذ وتحديد موقعه، مما يعزز الموثوقية في مجالات حيوية مثل التصوير الطبي والمراقبة الصناعية. غالبًا ما تكون الأساليب التقليدية عرضة للهجمات العدائية بسبب قيود بيانات التدريب، لكن PatchGuard تعالج هذه المشكلة من خلال دمج الشذوذ الزائف، مما يمثل تقدمًا كبيرًا في ضمان السلامة والدقة في هذه القطاعات الحيوية.

Un nuevo estudio presenta PatchGuard, un método innovador para la detección y localización de anomalías que mejora la fiabilidad en campos críticos como la imagen médica y el monitoreo industrial. Los enfoques tradicionales a menudo son vulnerables a ataques adversariales debido a la limitación de datos de entrenamiento, pero PatchGuard aborda este problema al incorporar pseudo-anomalías, lo que representa un avance significativo para garantizar la seguridad y precisión en estos sectores vitales.

Une nouvelle étude présente PatchGuard, une méthode révolutionnaire pour la détection et la localisation des anomalies qui améliore la fiabilité dans des domaines critiques comme l'imagerie médicale et la surveillance industrielle. Les approches traditionnelles sont souvent vulnérables aux attaques adversariales en raison de données d'entraînement limitées, mais PatchGuard résout ce problème en intégrant des pseudo-anomalies, ce qui représente une avancée significative pour garantir la sécurité et la précision dans ces secteurs vitaux.

A new study introduces PatchGuard, a groundbreaking method for anomaly detection and localization that enhances reliability in critical fields like medical imaging and industrial monitoring. Traditional approaches often struggle against adversarial attacks due to limited training data, but PatchGuard addresses this issue by incorporating pseudo anomalies, making it a significant advancement in ensuring safety and accuracy in these vital sectors.

PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies

arXiv:2511.16845v1 Announce Type: new 
Abstract: Ordinal classification has been widely applied in many high-stakes applications, e.g., medical imaging and diagnosis, where reliable uncertainty quantification (UQ) is essential for decision making. Conformal prediction (CP) is a general UQ framework that provides statistically valid guarantees, which is especially useful in practice. However, prior ordinal CP methods mainly focus on heuristic algorithms or restrictively require the underlying model to predict a unimodal distribution over ordinal labels. Consequently, they provide limited insight into coverage-efficiency trade-offs, or a model-agnostic and distribution-free nature favored by CP methods. To this end, we fill this gap by propose an ordinal-CP method that is model-agnostic and provides instance-level optimal prediction intervals. Specifically, we formulate conformal ordinal classification as a minimum-length covering problem at the instance level. To solve this problem, we develop a sliding-window algorithm that is optimal on each calibration data, with only a linear time complexity in K, the number of label candidates. The local optimality per instance further also improves predictive efficiency in expectation. Moreover, we propose a length-regularized variant that shrinks prediction set size while preserving coverage. Experiments on four benchmark datasets from diverse domains are conducted to demonstrate the significantly improved predictive efficiency of the proposed methods over baselines (by 15% decrease on average over four datasets).

تم اقتراح طريقة جديدة للتصنيف الترتيبي تركز على التنبؤ المتوافق (CP) لتحسين قياس عدم اليقين في التطبيقات الحساسة مثل التصوير الطبي والتشخيص. تهدف هذه الطريقة المستقلة عن النموذج إلى توفير فترات تنبؤ مثالية على مستوى الحالة، مما يعالج القيود التي كانت تعاني منها طرق CP الترتيبية السابقة التي اعتمدت على خوارزميات تجريبية أو توزيعات أحادية الشكل.

Se ha propuesto un nuevo método para la clasificación ordinal, centrado en la predicción conforme (CP) para mejorar la cuantificación de la incertidumbre en aplicaciones críticas como la imagenología médica y el diagnóstico. Este enfoque, independiente del modelo, tiene como objetivo proporcionar intervalos de predicción óptimos a nivel de instancia, abordando las limitaciones de los métodos CP ordinales anteriores que dependían de algoritmos heurísticos o distribuciones unimodales.

Une nouvelle méthode de classification ordinale a été proposée, axée sur la prédiction conforme (CP) pour améliorer la quantification de l'incertitude dans des applications critiques telles que l'imagerie médicale et le diagnostic. Cette approche, indépendante du modèle, vise à fournir des intervalles de prédiction optimaux au niveau des instances, répondant aux limitations des méthodes CP ordinales précédentes qui reposaient sur des algorithmes heuristiques ou des distributions unimodales.

A new method for ordinal classification has been proposed, focusing on conformal prediction (CP) to enhance uncertainty quantification in high-stakes applications like medical imaging and diagnosis. This model-agnostic approach aims to provide optimal prediction intervals at the instance level, addressing limitations of previous ordinal CP methods that relied on heuristic algorithms or unimodal distributions.

Provably Minimum-Length Conformal Prediction Sets for Ordinal Classification

<a href="https://www.techspot.com/news/110361-us-throwing-big-money-rare-earth-tech-taking.html" target="_blank"><img src="https://www.techspot.com/images2/news/ts3_thumbs/2025/11/2025-11-23-ts3_thumbs-794.jpg" width="800" height="560" style="padding: 15px 0" title="The US is throwing big money at rare-earth tech, taking on China's dominance with new chemistry" /></a> Rare-earth elements are embedded deep within much of today's hardware stack, but the real bottleneck is no longer geological reserves – it is the industrial circuitry that converts ore and scrap into high-purity oxides and finished magnets. Smartphones, high-torque electric-vehicle motors, direct-drive wind turbines, and advanced medical imaging all rely... <a href="https://www.techspot.com/news/110361-us-throwing-big-money-rare-earth-tech-taking.html">Read Entire Article</a>

تستثمر الولايات المتحدة بشكل كبير في تكنولوجيا العناصر الأرضية النادرة لتقليل هيمنة الصين في هذا القطاع، مع التركيز على العمليات الصناعية التي تحول الخام إلى مواد عالية النقاء ضرورية للتكنولوجيا الحديثة. تأتي هذه الخطوة في وقت تعتبر فيه العناصر الأرضية النادرة حيوية لمجموعة متنوعة من التطبيقات، بما في ذلك الهواتف الذكية والمركبات الكهربائية.

Estados Unidos está invirtiendo significativamente en tecnología de tierras raras para reducir la dominación de China en este sector, centrándose en los procesos industriales que convierten el mineral en materiales de alta pureza esenciales para la tecnología moderna. Este movimiento se produce en un momento en que los elementos de tierras raras son cruciales para diversas aplicaciones, incluidos los teléfonos inteligentes y los vehículos eléctricos.

Les États-Unis investissent massivement dans la technologie des terres rares pour réduire la domination de la Chine dans ce secteur, en se concentrant sur les processus industriels qui transforment le minerai en matériaux de haute pureté essentiels pour la technologie moderne. Ce mouvement intervient alors que les éléments des terres rares sont cruciaux pour diverses applications, y compris les smartphones et les véhicules électriques.

The United States is significantly investing in rare-earth technology to reduce China's dominance in the sector, focusing on the industrial processes that convert ore into high-purity materials essential for modern technology. This move comes as rare-earth elements are crucial for various applications, including smartphones and electric vehicles.

PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies

Was this article worth reading? Share it

GPTHumanizer

LangWatch

Blunge