In image anomaly detection, significant advancements have been made using un- and self-supervised methods with datasets containing only normal samples. However, these approaches often struggle with fine-grained anomalies. This paper introduces GRAD: Bi-Grid Reconstruction for Image Anomaly Detection, which employs two continuous grids to enhance anomaly detection from both normal and abnormal perspectives. In this work: 1) Grids as feature repositories that improve generalization and mitigate the Identical Shortcut (IS) issue; 2) An abnormal feature grid that refines normal feature boundaries, boosting detection of fine-grained defects; 3) The Feature Block Paste (FBP) module, which synthesizes various anomalies at the feature level for quick abnormal grid deployment. GRAD’s robust representation capabilities also allow it to handle multiple classes with a single model. Evaluations on datasets like MVTecAD, VisA, and GoodsAD show significant performance improvements in fine-grained anomaly detection. GRAD excels in overall accuracy and in discerning subtle differences, demonstrating its superiority over existing methods.

In the domain of image anomaly detection, significant progress has been made in unsupervised and self-supervised methods with datasets containing only normal samples. Although these methods perform well in general industrial anomaly detection scenarios, they often struggle with over- or under-detection when faced with fine-grained anomalies in products. In this paper, we propose GRAD: Bi-Grid Reconstruction for Image Anomaly Detection, which utilizes two continuous grids to detect anomalies from both normal and abnormal perspectives. In this work: 1) Grids serve as feature repositories to assist in the reconstruction task, achieving stronger generalization compared to discrete storage, while also helping to avoid the Identical Shortcut (IS) problem common in general reconstruction methods. 2) An additional grid storing abnormal features is introduced alongside the normal grid storing normal features, which refines the boundaries of normal features, thereby enhancing GRAD's detection performance for fine-grained defects. 3) The Feature Block Pasting (FBP) module is designed to synthesize a variety of anomalies at the feature level, enabling the rapid deployment of the abnormal grid. Additionally, benefiting from the powerful representation capabilities of grids, GRAD is suitable for a unified task setting, requiring only a single model to be trained for multiple classes. GRAD has been comprehensively tested on classic industrial datasets including MVTecAD, VisA, and the newest GoodsAD dataset, showing significant improvement over current state-of-the-art methods.

In this paper, we propose a security-enhanced integrated sensing and communication (ISAC) system enabled by phase-coupled Intelligent Omni-Surfaces (IOS). The IOS, without the need for additional sensors, divides the entire space into two parts: a communication part and a sensing part. The IOS serves multiple users by providing communication services while also creating a virtual line-of-sight (LOS) link for target sensing. Moreover, the IOS is utilized to ensure physical layer security. Due to the possibility of the sensing target intercepting communication information, it is crucial to limit the information leakage rate. To achieve this, we jointly design the communication beamformers, sensing beamformer, and the IOS’s two phase-shift matrices. This joint design maximizes the sensing beam pattern gain towards the target, while meeting the requirements of the minimum signal-to-interference-plus-noise ratio (SINR) for communications and the maximum information leakage to possible malicious targets. To address this problem, we propose two types of alternative algorithms. The first type utilizes an independent phase-controlling model, while the second incorporates a coupled-phase model. Both algorithms aim to optimize the IOS’s phase-shift matrices.

Reconfigurable Intelligent Surfaces (RISs) have been considered as a promising technology for next generation communications. The existing works mainly focused on the single-connected RIS, which is always mathematically characterized as diagonal phase shift matrices, because that each element of traditional RIS separately controls its status and would not like to apply influence to any other elements. Although there exist some research on sub-array RIS, the grouping conception is based on the controlling strategy level but not on the hardware realization level. Fortunately, a new type of beyond diagonal RIS (BD-RIS) is proposed with its mathematical expression is symmetric and not limited to traditional diagonal phase shift matrices. BD-RIS model can fully cover the previous various types of RIS like: reflective RIS, refractive RIS, IOS (Intelligent Omni-Surfaces) or STAR-RIS (Simultaneously Transmitting And Reflecting RIS). In this work, a new model of BD-RIS is proposed to explain its working principle. Then, the quantization strategy for the new model BD-RIS is analyzed. Finally, the optimization algorithms for BD-RIS in single-user multi-input-multi-output (SU-MISO) systems and in multi-user multi-input-multi-output (MU-MISO) systems are proposed. Simulation results verifies the effectiveness of the the algorithms’ performance.

Personalized recommender systems are becoming increasingly relevant and important in the study of polarization and bias, given their widespread use in filtering information spaces. Polarization is a social phenomenon, with serious consequences, in real-life, particularly on social media. Thus it is important to understand how machine learning algorithms, especially recommender systems, behave in polarized environments. In this paper, we study polarization within the context of the users' interactions with a space of items and how this affects recommender systems. We first formalize the concept of polarization based on item ratings and then relate it to the item reviews to investigate any potential correlation. We then propose a domain independent data science pipeline to automatically detect polarization using the ratings rather than the typical properties used to detect polarization, such as item's content or social network topology. We perform an extensive comparison of polarization measures on several benchmark data sets and show that our polarization detection framework can detect different degrees of polarization and outperforms existing measures in capturing an intuitive notion of polarization. Our work is an essential step toward quantifying and detecting polarization in ongoing ratings and in benchmark data sets, and to this end, we use our developed polarization detection pipeline to compute the polarization prevalence of several benchmark data sets. It is our hope that this work will contribute to supporting future research in the emerging topic of designing and studying the behavior of recommender systems in polarized environments.