Search Results (2,643)

The rapid evolution of malware families poses significant challenges for cybersecurity systems, particularly when newly emerging threats lack sufficient labeled data. Although image-based deep learning approaches have achieved strong performance under fully supervised conditions, their dependence on retraining limits adaptability in dynamic environments. To address this issue, we propose a Retrieval-Augmented Few-Shot Malware Detection Framework that integrates binary-to-image visualization, multimodal embedding using a frozen Vision–Language Model (Qwen2.5-VL), and similarity-based external memory retrieval. Malware binaries are converted into grayscale images and embedded into a semantic vector space without task-specific fine-tuning. During inference, query samples retrieve similar support embeddings from a vector database, and predictions are generated through similarity-weighted aggregation, enabling adaptation without parameter updates. Evaluated on the MalImg dataset with 25 malware families under 1-shot to 10-shot settings, the framework achieves 0.886 accuracy in the 10-shot configuration. Ablation results demonstrate that combining VLM embeddings with retrieval mechanisms provides consistent improvements over individual components. These findings highlight the effectiveness of decoupling representation learning from adaptation for scalable few-shot malware detection. Full article

The aim of this work is to explore a load frequency control (LFC) strategy in micro hydro power plants (MHPPs). Using MATLAB/Simulink, we examined several variants of genetic algorithms (GAs), including Roulette, Tournament, and Uniform, which are utilized to optimize tuning proportional integral derivative (PID) parameters by addressing the problem of instability caused by load variations. The performances are compared with conventional PID methods and other advanced techniques like particle swarm optimization (PSO), adaptive neuro-fuzzy inference system (ANFIS), and artificial neural networks (ANN) algorithms for both single and dual-area MHPP systems. The results show that the GA-optimized PID controller with the roulette wheel achieves the fastest settling time of 0.3 s and the smallest undershoot of 0.015 pu in the single area. Also, optimizing GA demonstrates superior performance in the dual area, with the fastest settling times of 2.5 s for both Roulette and Uniform. In contrast, PSO is slower than GA, and conventional PID requires a much longer settling time of 19.8 s, a similar result occurring in the dual area. These findings confirm the effectiveness of the GA-optimized PID controller, especially the Roulette variant, as a reliable and fast solution for maintaining frequency stability in MHPPs. Full article

The fuel consumption of power-split hybrid electric vehicles (PSHEVs) can be effectively reduced via mode transition that includes the engine process. However, factors such as engine torque ripple, system parameter uncertainties, and variations in torsional vibration characteristics can easily induce drivetrain vibration. These factors not only degrade ride comfort but also lead to a fundamental control challenge. The inherent trade-off between rapid response and stability is difficult to reconcile. In addition, the lack of adaptive mechanisms further limits consistent performance under varying conditions. To tackle these problems, a scenario-adaptive composite robust control (SACRC) strategy is proposed. The strategy consists of a UIO (unknown input observer)-based torque observation module, an adaptive VSS-LMS approach, and an H∞ controller with self-tuning parameters. Firstly, a six-degree-of-freedom dynamic model of the PSHEV transmission system is established with excitation sources, considering the characteristics of dual elastic elements. Secondly, a UIO-based torque observer is designed using a simplified dual-elastic-element model. By using engine speed and output shaft speed, the observer can accurately identify the torque transmitted by the torsional damper and drive shaft. Then, an adaptive VSS-LMS and H∞ controller with self-tuning parameters is constructed to ensure a balanced performance between fast torsional vibration suppression and control stability. Finally, simulation and experimental results demonstrate that the proposed strategy provides favorable adaptability to complex scenarios, and unifies the performance goals of rapidity, stability, and robustness. Full article

Continuous, non-invasive cardiopulmonary monitoring is receiving increasing attention as population aging and chronic diseases rise. Acoustic sensing provides diagnostically relevant information with relatively simple hardware. Yet, physiological body sounds span heterogeneous and partially overlapping spectra and are highly susceptible to environmental noise and motion artifacts, which limit conventional stethoscopes and fixed-frequency sensors. Frequency-Tunable Acoustic Sensors (FTAS) offer a promising route toward frequency-selective amplification and adaptive interference suppression by matching their resonance to target signals, thereby potentially supporting multi-site monitoring and personalized diagnostics on a single platform. This review starts with an overview of physiological sound generation and the evolution of auscultation, then surveys mainstream medical acoustic transducers (piezoelectric, capacitive microelectromechanical systems (MEMS), piezoresistive and triboelectric) and their limitations in frequency selectivity. Resonance-tuning strategies are classified into three paradigms: electrical tuning, material-based tuning, and geometric reconfiguration, and their tuning ranges, response characteristics, and representative implementations are comparatively discussed. Finally, this review discusses the potential translational value of FTAS in physiological acoustic signal monitoring, particularly in cardiovascular and respiratory assessment, and emphasizes the remaining challenges, including the trade-off between sensitivity and selectivity, as well as long-term biocompatibility. At the same time, this review highlights their development prospects in customizable acoustic sensing platforms. Full article

Thermochromic materials (TCMs) have attracted increasing attention as passive adaptive materials for solar regulation in buildings because they can reversibly change their optical properties in response to temperature without external energy input. Owing to this temperature-triggered optical modulation, they have been widely investigated for smart windows, temperature indicators, anti-counterfeiting labels, and flexible devices. In recent years, representative systems such as VO₂-based materials, polymers, hydrogels, and organic–inorganic hybrids have shown steady progress, especially in transition-temperature tuning, spectral selectivity, and cycling stability. This review summarizes the main classes of TCMs as well as their color-changing mechanisms, preparation methods, and performance-regulation strategies, with an emphasis on building energy efficiency and passive solar regulation. Typical applications and current bottlenecks are also discussed, including response speed, durability, environmental compatibility, and large-scale manufacturing. Finally, several practical directions for future work are highlighted, particularly low-cost synthesis, multifunctional integration, and application-oriented material design. Full article

With the increasing demand for enhanced security and surveillance, the integration of multimodal AI has shown significant promise. We developed and fine-tuned a transformer-based model, the Large Language and Vision Assistant–OneVision, tailored for human event and action recognition. By utilizing a multimodal approach, we identified specific human actions, including eating, running, fighting, sitting, and sleeping, within diverse real-world settings. Through knowledge distillation and Low-Rank Adaptation, the model’s performance was optimized in demonstrating substantial improvements in context-aware recognition and response generation. Evaluation results showed recall-oriented understudy for obtaining evaluation (ROUGE)-1 score of 0.6844, ROUGE-2 score of 0.5751, ROUGE-L score of 0.6520, and the bilingual evaluation understudy score of 68.20, demonstrating significant gains in accuracy and interpretability. The model’s success highlights its potential for real-time applications in surveillance, healthcare, and interactive AI systems, providing reliable, efficient, and context-sensitive human action detection. Full article

Short-term load forecasting is a key capability for smart-grid operation, but real smart-meter streams are affected by missing values, communication noise, and non-stationary consumption patterns. This paper studies forecasting using raw smart-meter data collected from domestic consumers in a medium-sized city in southern Spain. In particular, we assess Hierarchical Temporal Memory (HTM), a biologically inspired online sequence learner, against a family of Long Short-Term Memory (LSTM)-based recurrent baselines. HTM offers continual adaptation and avoids a separate training phase, whereas LSTM relies on offline supervised training and may require retraining or fine-tuning under distribution shift. For five-step-ahead forecasting, HTM achieved a test RMSE of 251 kWh (about 15% of average consumption). After hyperparameter optimization, the best tested LSTM configuration achieved a test RMSE of approximately 250 kWh under clean conditions, indicating nearly identical point accuracy between the two approaches. Under synthetic Gaussian-noise injection, however, HTM remained comparatively stable, whereas the optimized LSTM configuration degraded markedly under the tested perturbation protocol. In addition, HTM exhibited a lower runtime in the tested CPU-based implementation. These findings suggest that HTM is a viable online alternative for aggregated smart-meter forecasting, offering competitive accuracy together with a favorable operational profile under the specific evaluation setup considered here. Full article

Automatic modulation recognition (AMR) is increasingly relevant to communication-sensing front ends in robotic and human–robot collaborative systems, where reliable spectrum awareness and adaptive wireless reception are desired. However, existing methods often degrade sharply at low signal-to-noise ratios (SNRs), and large language models (LLMs) are not natively compatible with continuous I/Q signals due to the inherent modality gap. We propose BioLAMR, a GPT-2 adaptation framework for AMR inspired by the auditory system’s parallel time–frequency processing and cortical hierarchy. The framework combines bio-inspired dual-domain feature extraction with parameter-efficient LLM adaptation. BioLAMR includes three components. First, a lightweight dual-domain fusion (LDDF) module extracts complementary time- and frequency-domain features and fuses them through channel and spatial attention. Second, a convolutional embedding module converts continuous I/Q signals into GPT-2-compatible sequences without discrete tokenization. Third, a hierarchical fine-tuning strategy updates only 8.9% of parameters to preserve pretrained knowledge while adapting to modulation recognition. Experiments on the RadioML2016.10a and RadioML2016.10b benchmarks show that BioLAMR achieves overall accuracies of 64.99% and 67.43%, outperforming the strongest competing method by 2.60 and 2.47 percentage points, respectively. Under low-SNR conditions, it reaches 36.78% and 38.14%, the best results among the compared methods. Ablation studies verify the contribution of each component. These results demonstrate that combining dual-domain signal modeling with parameter-efficient GPT-2 adaptation is an effective route to robust AMR in challenging wireless environments. Full article

Formation control of unmanned surface vehicles (USVs) in complex marine environments is required to contend with strongly coupled, high-dimensional disturbances. A Multi-Agent Active Disturbance Rejection Control (MAADRC) framework is developed for this purpose. The design centers on a distributed extended state observer (DESO) coupled with a dual-channel feedback structure—NEFL-GCO and LGL-FC—that collectively maintains formation geometry. Three main ideas underpin the approach. First, a bandwidth-efficient distributed observation scheme enables agents to share disturbance estimates while using substantially less communication bandwidth. Second, an adaptive consensus compensation mechanism accommodates parameter variations as formations evolve. Third, a formation-compatible obstacle avoidance algorithm enhances reliability in congested waters. To evaluate the control structure and optimize its parameters, a multi-agent reinforcement learning (MARL) method—specifically Multi-Agent Proximal Policy Optimization (MAPPO)—is employed. The MARL agent tunes two critical parameters: observer bandwidth and nonlinear feedback gain, thereby establishing a performance baseline. After ten million training steps, the MAPPO-optimized MAADRC achieves a tracking root-mean-square error (RMSE) of 1.18 m. This value lies within 3% of the manually tuned result of 1.21 m, indicating that the bandwidth parameterization is near-optimal. Extensive simulations incorporating realistic wind, wave and current disturbances demonstrate a dynamic obstacle avoidance success rate maintaining an expected level, alongside consistently low formation tracking errors. Collectively, these findings confirm the resilience and practical utility of the proposed framework in demanding maritime settings. Full article

Deep learning has revolutionized medical image segmentation; however, the clinical deployment of state-of-the-art models is severely impeded by their quadratic computational complexity and substantial resource demands, particularly in multisensor and multimodal imaging scenarios. In response, the field is undergoing a paradigm shift towards efficiency, characterized by the rise of linear-complexity architectures and the optimization of foundation models. This paper presents a comprehensive survey of efficient medical image segmentation methodologies, systematically reviewing the evolution from heavy, accuracy-driven models to lightweight, deployment-ready paradigms. In particular, we highlight the growing importance of efficient segmentation in multisensor medical imaging, where heterogeneous data sources such as CT, MRI, ultrasound, and infrared imaging introduce additional challenges in scalability and computational cost. We propose a novel taxonomy that categorizes these advancements into four distinct streams: (1) Mamba and State Space Models, which leverage selective scanning mechanisms to achieve global receptive fields with linear complexity; (2) Efficient Adaptation of Foundation Models, focusing on parameter-efficient fine-tuning and knowledge distillation to tailor the Segment Anything Model (SAM) for medical domains; (3) Advanced Lightweight Architectures, covering the resurgence of large-kernel CNNs and the emergence of Kolmogorov–Arnold Networks (KANs); and (4) Data-Efficient Strategies, including semi-supervised and federated learning to address annotation scarcity. Furthermore, we conduct a rigorous comparative analysis of representative algorithms on mainstream benchmarks, providing a granular evaluation of the trade-offs between segmentation accuracy and computational overhead. The survey also discusses key challenges in multisensor and multimodal settings, including modality heterogeneity, data fusion complexity, and resource constraints. Finally, we identify critical challenges and outline future research directions, serving as a roadmap for the development of next-generation efficient and scalable medical image analysis systems. Full article

Real-time coordination across OEM–Tier-1 manufacturing networks remains challenging due to delayed shop-floor data, stochastic machine availability, and the need for schedule stability. This paper presents a protocol-agnostic, two-stage mixed-integer linear programming (MILP) framework for real-time family-level scheduling. The method integrates MTConnect-like data streams without requiring adherence to any single communication standard. In Stage 1, a baseline plan is generated using expected capacity; in Stage 2, a rolling-horizon recourse model adapts the plan to observed (possibly lagged) capacity while incorporating a stability penalty to control resequencing. A synthetic OEM–Tier-1 testbed with three machines (two Tier-1, one OEM) is used to benchmark performance under real-time (L = 0) and delayed (L = 5) data scenarios. Across these scenarios, the real-time rolling scheduler improves strict on-time fulfillment by approximately 70% and eliminates terminal backlog relative to static planning, while MILP solve times remain under 0.1 s per cycle. Sensitivity experiments that vary disruption intensity, replanning interval (Δ), and stability weight (λ) show consistent qualitative trends and illustrate how the framework can be tuned to balance service performance against schedule stability without sacrificing computational tractability. Full article

Hyperspectral unmixing aims to estimate endmember spectra and their corresponding abundance fractions at the subpixel scale, which is a critical preprocessing step for quantitative analysis of hyperspectral remote sensing imagery. While deep learning-based methods have achieved remarkable progress, three fundamental challenges remain: (i) reliance on a single shared spatial prior that cannot decouple the heterogeneous spatial patterns of different land covers; (ii) the lack of synergy in jointly optimizing endmember extraction and abundance estimation; (iii) the poor robustness of unsupervised training to complex mixtures, noise, and class imbalance. To address these issues, we propose a novel unsupervised unmixing framework that integrates adaptive orthogonal multi-faceted graph representation with curriculum learning. Specifically, we design an Adaptive Orthogonal Multi-Faceted Graph Generator (AOMFG) to learn a set of independent orthogonal graph structures, achieving spatially informed decoupling of land cover patterns. Then, a dual-branch collaborative optimization network is constructed: a Graph Convolutional Network (GCN) branch that incorporates the learned spatial topological priors for abundance estimation, and a 1D Convolutional Neural Network (1DCNN) branch that employs a query-attention mechanism to adaptively aggregate pure spectral features for endmember extraction. Finally, we introduce a three-stage curriculum learning strategy that progressively fine-tunes the model, which significantly enhances its performance. Extensive experiments on three widely used real-world benchmark datasets demonstrate that our proposed framework consistently outperforms state-of-the-art methods in both endmember extraction and abundance estimation accuracy. Comprehensive ablation studies, parameter sensitivity analysis, and noise robustness tests further validate the effectiveness of each core component. Full article

Dialectal variation presents a major challenge for deploying medical language models in real-world healthcare settings, where patient–clinician communication often occurs in regional vernaculars rather than standardized language forms. This challenge is particularly pronounced in the Arabic-speaking world, where clinical interactions frequently take place in diverse dialects that differ substantially from Modern Standard Arabic. Fine-tuning and maintaining separate models for each dialect is computationally inefficient and difficult to scale, motivating more integrated approaches. In this work, we present MENARA, an Arabic medical language model constructed by merging Egyptian Arabic, Moroccan Darija, and medical-domain specialists through model merging. We extend prior feasibility findings through comprehensive evaluation of cross-dialect performance, medical safety, and cross-lingual knowledge retention. Specifically, we introduce a fine-grained dialect composition analysis to quantify lexical purity and structured code-switching behavior, benchmark against state-of-the-art Arabic LLMs, conduct subject-matter-expert assessment of both dialectal fidelity and medical appropriateness. The results show that model merging preserves core medical competence while enabling robust dialectal adaptation, achieving strong cross-dialect fidelity while substantially reducing storage and deployment overhead compared to maintaining separate models. These findings establish model merging as a potentially practical and resource-efficient paradigm for dialect-aware medical NLP in linguistically fragmented healthcare environments. Full article

Software Development (Dev) and Information Technology Operations (Ops) rely on different process parameters, such as robustness, resilience, and in-line optimizations. Resilience is a key requirement when adopting various enterprise features to ensure system stability and swift recovery from failures. Based on different enterprises’ adoption conditions, the need for resilience and its optimization is to be designed. In this article, a low-complexity, adoptable, and validated resilience model is proposed to improve the efficiency of the DevOps functional phase. This proposed model uses intrinsic deep lift features of the applications to assess its resilience. In this case, optimization is performed using trust, robustness, and resilience parameters, as per the application’s demand. A fine-to-coarse tuning strategy is applied to both individual and permuted parameters to improve the adaptability and scalability of the DevOps implementation. Considering parameter permutations, selective tuning is also feasible through deep lift learning to improve resilience with reduced complexity. This model is efficient in leveraging adoptability, achieving 12.43% for the plan and 13.17% for the feedback phases, for maximum execution time. Full article

Full fine-tuning of pre-trained models sometimes requires inserting trainable layers into the middle of a pre-trained backbone, but such middle-layer insertion can severely degrade downstream performance. We hypothesize that this degradation arises because conventionally inserted layers, when randomly initialized and combined with output-side activation, perturb intermediate representations before the pre-trained model has adapted. We study this phenomenon across natural language processing and computer vision benchmarks by varying insertion locations, the number of inserted layers, and activation designs. To address this problem, we propose a practical stabilization method for middle-layer insertion under full fine-tuning: a bias-free inserted layer with unit initialization and weight-side activation. This design is intended to remain closer to an identity-like transformation at initialization, thereby reducing initialization-time perturbation rather than claiming exact preservation of the original representations. In the tested DeBERTa-v3, T5-base, and ViT-base settings, the proposed method substantially mitigates the severe degradation caused by naive middle-layer insertion and maintains performance close to the no-added-layer baseline, including settings with up to 24 inserted layers. Full article