Search Results (760)

Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention (DMA) mechanism, enabling effective cross-modal knowledge transfer for capturing macro- and micro-anomalies. A Stabilized Attention-based Pooling (SAP) module adaptively aggregates discriminative representations using self-generated anomaly heatmaps, enhancing localization accuracy and mitigating feature dilution. Trained solely in auxiliary datasets with multi-task segmentation and contrastive losses, the approach requires no target-domain samples. Extensive evaluation across seven benchmarks (MVTec AD, VisA, BTAD, MPDD, KSDD, DAGM, DTD-Synthetic) demonstrates state-of-the-art performance, achieving 93.4% image-level AUROC, 94.3% AP, 96.9% pixel-level AUROC, and 92.4% AUPRO on average. Ablation studies confirm the efficacy of DMA and SAP, while qualitative results highlight superior boundary precision and noise suppression. The framework offers a scalable, annotation-efficient solution for real-world industrial anomaly detection. Full article

Transfer learning, which utilizes extensive simulation data to overcome the limitations of scarce and expensive experimental data, has emerged as a powerful approach for predictive modeling in various physical domains. This study presents a comprehensive framework to improve the predictive performance of transfer learning, focusing on quasi-zero stiffness (QZS) systems with limited experimental datasets. The proposed framework systematically examines the interplay among three critical factors in the target domain: data augmentation, layer-freezing configurations, and neural network architecture. Simulation-driven synthetic data are generated to capture dynamic features not represented in the sparse experimental data. The optimal transfer depth is explored by evaluating different scenarios of selective layer freezing and fine-tuning. Results show that partial transfer strategies outperform both full-transfer and non-transfer approaches, leading to more stable and accurate predictions. To investigate hierarchical transfer, both symmetric and asymmetric network architectures are designed, embedding physically meaningful representations from simulations into the deeper layers of the target model. Furthermore, an attention mechanism is integrated to emphasize material-specific characteristics. Building on these components, the proposed simulation-driven framework predicts the full force–displacement responses of QZS systems using only 12 experimental samples. Through a systematic comparison of three datasets (direct transfer, linear correction, FEM-based correction), three network architectures, and seven layer-freezing scenarios, the framework achieves a best test performance of R² = 0.978 and MAE = 0.34 Newtons. Full article

This research addresses the critical need for efficient image annotation in precision agriculture, using the wild blueberry (Vaccinium angustifolium Ait.) cropping system as a representative application to enable data-driven crop management. Tasks such as automated berry ripeness detection, plant disease identification, plant growth stage monitoring, and weed detection rely on extensive annotated datasets. However, manual annotation is labor-intensive, time-consuming, and impractical for large-scale agricultural systems. To address this challenge, this study evaluates an automated annotation pipeline that integrates zero-shot detection models from two frameworks (Grounding DINO and YOLO-World) with the Segment Anything Model version 2 (SAM2). The models were tested on detecting and segmenting ripe wild blueberries, developmental wild blueberry buds, hair fescue (Festuca filiformis Pourr.), and red leaf disease (Exobasidium vaccinii). Grounding DINO consistently outperformed YOLO-World, with its Swin-T achieving mean Intersection over Union (mIoU) scores of 0.694 ± 0.175 for fescue grass and 0.905 ± 0.114 for red leaf disease when paired with SAM2-Large. For ripe wild blueberry detection, Swin-B with SAM2-Small achieved the highest performance (mIoU of 0.738 ± 0.189). Whereas for wild blueberry buds, Swin-B with SAM2-Large yielded the highest performance (0.751 ± 0.154). Processing times were also evaluated, with SAM2-Tiny, Small, and Base demonstrating the shortest durations when paired with Swin-T (0.30–0.33 s) and Swin-B (0.35–0.38 s). SAM2-Large, despite higher segmentation accuracy, had significantly longer processing times (significance level α = 0.05), making it less practical for real-time applications. This research offers a scalable solution for rapid, accurate annotation of agricultural images, improving targeted crop management. Future research should optimize these models for different cropping systems, such as orchard-based agriculture, row crops, and greenhouse farming, and expand their application to diverse crops to validate their generalizability. Full article

We present a detailed derivation of the quantum and quantum–thermal effective action for non-relativistic systems, starting from the single-particle case and extending to the Gross–Pitaevskii (GP) field theory for weakly interacting bosons. In the single-particle framework, we introduce the one-particle-irreducible (1PI) effective action formalism, taking explicitly into account the choice of the initial quantum state, its saddle-point plus Gaussian-fluctuation approximation, and its finite-temperature extension via Matsubara summation, yielding a clear physical interpretation in terms of zero-point and thermal contributions to the Helmholtz free energy. The formalism is then applied to the GP action, producing the 1PI effective potential at zero and finite temperature, including beyond-mean-field Lee–Huang–Yang and thermal corrections. We discuss the gapless and gapped Bogoliubov spectra, their relevance to equilibrium and non-equilibrium regimes, and the role of regularization. Applications include the inclusion of an external potential within the local density approximation, the derivation of finite-temperature Josephson equations, and the extension to D-dimensional systems, with particular attention to the zero-dimensional limit. This unified approach provides a transparent connection between microscopic quantum fluctuations and effective macroscopic equations of motion for Bose–Einstein condensates. Full article

Continual learning for open-vocabulary object detection aims to enable pretrained vision–language detectors to adapt to diverse specialized domains while preserving their zero-shot generalization capabilities. However, existing methods primarily focus on mitigating catastrophic forgetting, often neglecting the substantial domain shifts commonly encountered in real-world applications. To address this critical oversight, we pioneer Open-Domain Continual Object Detection (OD-COD), a new paradigm that requires detectors to continually adapt across domains with significant stylistic gaps. We propose Disentangled Domain Knowledge-Aided Learning (D-Know) to tackle this challenge. This framework explicitly disentangles domain-general priors from category-specific adaptation, managing them dynamically in a scalable domain knowledge base. Specifically, D-Know first learns domain priors in a self-supervised manner and then leverages these priors to facilitate category-specific adaptation within each domain. To rigorously evaluate this task, we construct OD-CODB, the first dedicated benchmark spanning six domains with substantial visual variations. Extensive experiments demonstrate that D-Know achieves superior performance, surpassing current state-of-the-art methods by an average of 4.2% mAP under open-domain continual settings while maintaining strong zero-shot generalization. Furthermore, experiments under the few-shot setting confirm D-Know’s superior data efficiency. Full article

This study explores the decarbonization of the district heating network in Riva del Garda. The existing system (baseline) was modeled in EnergyPLAN, and future configurations were optimized using a Multi-Objective Evolutionary Algorithm (MOEA) to minimize both CO₂ emissions and annual costs. Nine decision variables were assessed under defined boundary conditions to generate alternative future scenarios grouped into five types. In Type A, a large deep geothermal cogeneration plant combined with a small biomass boiler achieved the only zero-emission solution, with lower annual costs than the baseline but high capital needs. Excluding deep geothermal cogeneration (Type B) led to dominance of the biomass boiler and waste heat recovery from the Alto Garda Power (AGP) plant; full decarbonization remained possible only with extensive biomass use at a higher cost. Removing biomass (Type C), the solar thermal plant, and the shallow geothermal heat pump enabled deep but costly decarbonization, including grid electricity dependence. Types D and E, dominated, respectively, by shallow geothermal heat pump and electric boiler, provided moderate emission reductions and further increase in costs. Across all types, thermal storage improved operational flexibility. These analyses were also extended to assess potential district heating network expansions within Riva del Garda and into the neighboring municipality of Arco. Full article

Vacuum circuit breakers (VCBs) have been extensively employed in switching shunt capacitor banks. However, research on the prestrike characteristics of double-break VCBs in making power frequency voltage remains limited. This study aims to investigate the influence of different closing time differences on the prestrike characteristics of double-break VCBs in making power frequency voltage, and to compare these influences with those of single-break VCBs. Experiments were conducted using vacuum interrupters rated at 24 kV, with contacts made of CuCr40 alloy doped with 1 wt% graphene. Taking the closing time of the high-voltage break as the time zero point, three closing time differences (0 ms, 0.727 ms, and −0.347 ms) were set, and experiments were carried out at six closing phase angles (from 0° to 150° in 30° increments) for each condition. The experimental results demonstrate that when the closing of the high-voltage break lags behind that of the low-voltage break by 0.347 ms, the double-break VCB exhibits optimal prestrike performance, where prestrike is almost entirely suppressed except at the 90° phase angle. Furthermore, the prestrike performance during the closing of the double-break VCB is significantly superior to that of the single-break VCB, characterized by a steeper RDDS curve. These findings provide a theoretical basis for the design of control-switching double-break VCBs. Full article

We briefly review the early development of the mean-field dynamics for cooperatively interacting quantum many-body systems, mapped to pseudo-spin (Ising-like) systems. We start with (Anderson, 1958) pseudo-spin mapping the BCS (1957) Hamiltonian of superconductivity, reducing it to a mean-field Hamiltonian of the XY (or effectively Ising) model in a transverse field. Then, we obtain the mean-field estimate for the equilibrium gap in the ground-state energy at different temperatures (gap disappearing at the transition temperature), which fits Landau’s (1949) phenomenological theory of superfluidity. We then present in detail a general dynamical extension (for non-equilibrium cases) of the mean-field theory of quantum Ising systems (in a transverse field), following de Gennes’ (1963) decomposition of the mean field into the orthogonal classical cooperative (longitudinal) component and the quantum (transverse) component, with each of the component following Suzuki–Kubo (1968) mean-field dynamics. Next, we discuss its applications to quantum hysteresis in Ising magnets (in the presence of an oscillating transverse field), to quantum heat engines (employing the transverse Ising model as a working fluid), and to the quantum annealing of the Sherrington–Kirkpatrick (1975) spin glass by tuning down (to zero) the transverse field, which provides us with a very fast computational algorithm, leading to ground-state energy values converging to the best-known analytic estimate for the model. Finally, we summarize the main results obtained and draw conclusions about the effectiveness of the de Gennes–Suzuki–Kubo mean-field equations for the study of various dynamical aspects of quantum condensed matter systems. Full article

Common Weakness Enumerations (CWEs) and Common Vulnerabilities and Exposures (CVEs) are open knowledge bases that provide definitions, descriptions, and samples of code vulnerabilities. The combination of Large Language Models (LLMs) with vulnerability knowledge bases helps to enhance and automate code vulnerability repair. Several key factors come into play in this setting, including (1) the retrieval of the most relevant context to a specific vulnerable code snippet; (2) augmenting LLM prompts with the retrieved context; and (3) the generated artifact form, such as a code repair with natural language explanations or a code repair only. Artifacts produced by these factors often lack transparency and explainability regarding the rationale behind the repair. In this paper, we propose an LLM-enabled framework for explainable recommendation of vulnerable code repairs with techniques addressing each factor. Our method is data-driven, which means the data characteristics of the selected CWE and CVE datasets and the knowledge base determine the best retrieval strategies. Across 100 experiments, we observe the inadequacy of the SOTA metrics to differentiate between low-quality and irrelevant repairs. To address this limitation, we design the LLM-as-a-Judge framework to enhance the robustness of recommendation assessments. Compared to baselines from prior works, as well as using static code analysis and LLMs in zero-shot, our findings highlight that multifaceted LLMs guided by retrieval context produce explainable and reliable recommendations under a small to mild level of self-alignment bias. Our work is developed on open-source knowledge bases and models, which makes it reproducible and extensible to new datasets and retrieval strategies. Full article

Unmanned Aerial Vehicles (UAVs) are increasingly being used in critical infrastructure, defense, and civilian applications, and face new cybersecurity threats. In this work, we present a novel hybrid deep learning architecture that combines Mamba, Kolmogorov-Arnold Networks (KAN), and Liquid Neural Networks for real-time cyberattack detection in UAV systems. The proposed Mamba-KAN-Liquid (MKL) model integrates Mamba’s selective state-space mechanism for temporal dependency modeling, KAN’s learnable activation functions for feature representation, and Liquid networks’ dynamic adaptation capabilities for real-time anomaly detection. Extensive evaluations on CIC-IDS2017, CSE-CIC-IDS2018, and synthetic UAV telemetry datasets demonstrate that our model achieves detection rates exceeding 95% across six different attack scenarios, including GPS spoofing (97.3%), network jamming (95.8%), man-in-the-middle attacks (96.2%), sensor manipulation (94.7%), DDoS (98.1%), and zero-day attacks (89.4%). The model meets real-time processing requirements with an average inference time of 47.3 ms for a sample batch size of 32, making it suitable for practical deployment on resource-constrained UAV platforms. Full article

We investigate algebraic and topological properties of subsymmetric polynomials on finite- and infinite-dimensional spaces. In particular, we focus on the problem of the existence of an algebraic basis in the algebra of subsymmetric polynomials, as well as possible extensions of subsymmetric polynomials and analytic functions to larger spaces. We consider algebras of subsymmetric analytic functions of bounded type and their spectra, and study linear subspaces in the zero-sets of subsymmetric polynomials, as well as subspaces where a subsymmetric polynomial is symmetric. In addition, we propose some possible applications of subsymmetric polynomials in cryptography and in operator theory. Full article

Zero-shot action recognition remains challenging due to the visual–semantic gap and the persistent bias toward seen classes, particularly under the generalized setting where both seen and unseen categories appear during inference. To address these issues, we propose Multi-Scale Semantic Alignment framework for Zero-Shot Sports Action Recognition (MSA-ZSAR), a framework that integrates a multi-scale spatiotemporal feature extractor to capture both coarse and fine-grained motion dynamics, a dual-branch semantic alignment strategy that adapts to different levels of semantic availability, and a bias-suppression mechanism to improve the balance between seen and unseen recognition. This design ensures that the model can effectively align visual features with semantic representations while alleviating overfitting to source classes. Extensive experiments demonstrate the effectiveness of the proposed framework. MSA-ZSAR achieves 52.8% unseen accuracy, 69.7% seen accuracy, and 61.3% harmonic mean, consistently surpassing prior approaches. These results confirm that the proposed framework delivers balanced and superior performance in realistic generalized zero-shot scenarios. Full article

Open-vocabulary semantic segmentation (OVSS) is of critical importance for unmanned aerial vehicle (UAV) imagery, as UAV scenes are highly dynamic and characterized by diverse, unpredictable object categories. Current OVSS approaches mainly rely on the zero-shot capabilities of vision–language models (VLMs), but their image-level pretraining objectives yield ambiguous spatial relationships and coarse-grained feature representations, resulting in suboptimal performance in UAV scenes. In this work, we propose a novel hybrid framework for OVSS in UAV imagery, named HOSU, which leverages the priors of vision foundation models to unleash the potential of vision–language models in representing complex spatial distributions and capturing fine-grained small-object details in UAV scenes. Specifically, we propose a distribution-aware fine-tuning method that aligns CLIP with DINOv2 across intra- and inter-region feature distributions, enhancing the capacity of CLIP to model complex scene semantics and capture fine-grained details critical for UAV imagery. Meanwhile, we propose a text-guided multi-level regularization mechanism that leverages the text embeddings of CLIP to impose semantic constraints on the visual features, preventing their drift from the original semantic space during fine-tuning and ensuring stable vision–language correspondence. Finally, to address the pervasive occlusion in UAV imagery, we propose a mask-based feature consistency strategy that enables the model to learn stable representations, remaining robust against viewpoint-induced occlusions. Extensive experiments across four training settings on six UAV datasets demonstrate that our approach consistently achieves state-of-the-art performance compared with previous methods, while comprehensive ablation studies and analyses further validate its effectiveness. Full article

Robot-based rehabilitation emerges as a promise to enhance mobility and improve the rehabilitation outcomes in children with cerebral palsy. The study aimed to evaluate the preliminary effects of a robot-based therapy program with Atlas-2030 on spatiotemporal gait parameters, pelvis kinematics, gross-motor function, quality of life, and joint range-of-motion in children with cerebral palsy receiving care at a specialized rehabilitation center. This is a single-arm, institution-based, quantitative, longitudinal, pilot study with repeated measures. Sixteen sessions of a robot-based therapy program with the Atlas-2030 wearable exoskeleton were applied to all the children from APAC-IAP in Mexico City with cerebral palsy. Pre-intervention, after eight and sixteen sessions, the GMFM-66, the CP QoL-Child, and gait analysis were performed. The results suggest that an Atlas-2030 robot-based therapy program combined with therapeutic stimulation exhibited better scores on the modified Ashworth scale: hip flexors and extensors: 2.0(1.0), knee flexors and extensors: 2.0(2.9), p > 0.0167, and experience enhanced range of motion in hip flexion: 122.5(5) deg, and extension: 11(5) deg and knee extension: 0(5) deg, p < 0.0167, pelvis rotation approached zero on both sides (left: −1.99(14.04, right: 2.22(13.43), p > 0.0167) reducing the difference in laterality, inducing physiological muscle activation patterns, and higher scores in quality of life regarding well-being and acceptance: 17(1.0) and emotional well-being and self-esteem: 14.5 (1.0), p > 0.0167. The limitations of this study include the following: recruitment from a single specialty care center, the absence of a control group, and the adjusted significance level of p < 0.0167, which may lead to false negatives. Full article

This paper presents an innovative method for designing finite impulse response (FIR) filters. The method optimizes the desired frequency response attributes while simultaneously increasing the sparsity of the filter coefficients. Traditional FIR filter design techniques, such as the window method (FirW) and the Parks–McClellan (FirPM) algorithm, excel in meeting precise frequency-domain requirements but often result in dense impulse responses. In scenarios with limited resources, a sparse filter, which has numerous zero or nearly zero coefficients, has advantages such as decreased computational complexity, lower power consumption, and simplified hardware integration. The proposed (L₁/L₂ regularization) approach defines filter design as an iterative optimization challenge that decreases a composite objective function. This function combines an error term based on the L₂-norm to measure deviation from the target frequency response and an L₁-norm-based regularization term to encourage coefficient sparsity. By adjusting the regularization parameter λ, users can balance performance in the frequency domain with the level of impulse response sparsity. Extensive simulations reveal that compared with filters designed using FirW and FirPM, this method produces filters with competitive frequency characteristics while achieving significantly higher sparsity. This finding highlights its considerable potential for effective hardware and software implementation. The proposed FIR filter design method presents a compelling alternative to conventional paradigms, particularly for applications where implementation efficiency is a critical design constraint. Full article