Search Results (19,408)

Existing deep-learning-based motor imagery (MI) electroencephalogram (EEG) decoding methods face challenges in generalizing across sessions and providing channel-level physiological interpretability. These limitations hinder the practical application of MI-EEG systems. Accordingly, DO-PI-EATCNet (Dream-Optimization-Enhanced, Physics-Inspired, Efficient-Attention Temporal Channel Network) is proposed to improve generalization and interpretability in MI-EEG classification. Unlike models that simply combine multiple components, DO-PI-EATCNet assigns distinct roles to feature representation, temporal channel modeling, temporal regularization, and channel compactness. Latent-Projected Attention (LPA) enhances spatiotemporal discriminability by aligning attention in a low-dimensional latent space, and Temporal Channel Cascaded Collaborative Attention (TCCA) refines dependencies between time and channels. Fractional-Order Difference Temporal Consistency Loss (FD-TCL) is introduced as a neurodynamics-inspired temporal regularizer to reduce high-frequency fluctuations in prediction sequences and improve within-subject cross-session prediction stability. The Multi-Population Dream Optimization Algorithm (MPDOA) is used for channel selection to obtain a compact EEG channel subset and reduce computational load, although it introduces a slight accuracy decrease compared with the uncompressed full model. Under a within-subject cross-session protocol on the BCI Competition IV-2a four-class MI dataset, the final compact model achieves an average accuracy of 84.4% and Cohen’s κ of 0.790, outperforming the reimplemented baselines. Compared with the uncompressed LPA-TCCA-FD-TCL variant, MPDOA slightly decreases accuracy from 84.9% to 84.4%, but reduces EEG channels from 22 to about 15 and decreases MACs by 27%. Scalp topographies and selected-channel visualizations provide qualitative support for channel-level anatomical plausibility, as the selected electrodes are mainly located over expected sensorimotor-related regions, while t-SNE offers a descriptive visualization of the learned feature distributions. Full article

Large language models (LLMs) remain vulnerable to jailbreak attacks, especially in black-box settings where target-model gradients and internal tokenization are inaccessible. Recent information bottleneck-based defenses cast prompt protection as a compression problem, but existing methods still rely heavily on white-box optimization and the intrinsic alignment strength of the protected model. To address these limitations, we propose RIB-Guard, a safety-aware information bottleneck defense for black-box LLMs. RIB-Guard learns a token-level masking policy that extracts a minimally safety-sufficient prompt via reinforcement learning using only black-box feedback. In addition, it introduces an independent lightweight safety head to estimate residual jailbreak risk and provide model-agnostic safety guidance during training. The proposed framework jointly balances prompt compactness, benign utility preservation, and residual risk suppression within a unified objective. Experimental results on direct single-turn harmful and benign prompt settings show that RIB-Guard improves jailbreak robustness while maintaining competitive benign utility. By extending information bottleneck-based prompt protection from white-box to black-box settings, RIB-Guard provides a step toward safety-aware information-theoretic front-end defense for black-box LLMs. Full article

Skeletal kinetic mechanisms are essential for reducing the computational cost of ammonia combustion simulations while retaining the key chemical features governing ignition, flame propagation, and NO formation. This study extends the DRG-CSP-ANN reduction and optimization framework to ammonia combustion over a broader multi-condition parameter space, aiming to develop a compact skeletal mechanism applicable to different pressures, equivalence ratios, and temperatures. Sixteen detailed ammonia combustion mechanisms were first assessed against experimental data covering ignition delay time, laminar flame speed, and NOx species concentrations over wide ranges of pressure, temperature, equivalence ratio, and oxidizer composition. Based on the overall error evaluation, the detailed mechanism with the most balanced predictive performance was selected as the parent mechanism. The parent mechanism was then reduced using the Directed Relation Graph and Computational Singular Perturbation methods, yielding an initial skeletal mechanism, RA-Ori, with 20 species and 76 reactions. To compensate for the accuracy loss caused by mechanism reduction, an Artificial Neural Network surrogate was constructed to optimize the pre-exponential factors of selected sensitive reactions within their evaluated uncertainty ranges, leading to the final mechanism, RA-ANN. The validation results show that RA-ANN reasonably reproduces ignition delay times, laminar flame speeds, and NO concentrations under different ammonia combustion conditions. Quantitatively, RA-ANN reduces the overall error from 0.335 for RA-Ori to 0.206, corresponding to a 38.4% reduction, while maintaining the same compact size. Its overall error is close to that of the parent detailed mechanism and lower than that of several existing skeletal mechanisms considered in this work. These results demonstrate that the proposed DRG-CSP-ANN strategy can construct a compact ammonia skeletal mechanism that achieves a favorable balance between computational efficiency, predictive accuracy, and applicability over representative multi-condition ammonia combustion regimes. Full article

Vernacular settlements in hot–humid regions preserve climate-responsive spatial knowledge, yet evidence on how linked outdoor, transitional, and indoor spaces jointly shape microclimate and thermal comfort remains limited. This study investigates a compact Hakka settlement in southern Jiangxi, China, by integrating field measurements, calibrated simulation, PET-based thermal-comfort assessment, and parametric scenario comparison to examine microclimatic differentiation across cold alleys, patios, halls, semi-open interfaces, and interior rooms. The results reveal clear microclimatic gradients across the linked vernacular spatial sequence. During the summer afternoon peak, cold alleys reduced air temperature by approximately 2.5 °C and PET by approximately 8.5 °C relative to ordinary streets, while semi-enclosed spaces adjacent to patios reduced air temperature by approximately 4.0 °C but increased relative humidity by 8–12%, indicating a cooling–moisture trade-off. Measured and simulated air temperature and wind speed showed satisfactory agreement and reproduced the main thermal and ventilation hierarchy across the connected spaces. Parametric comparison further identified case-based geometry-performance tendencies under the tested boundary conditions: within the tested cold-alley scenarios, widths of approximately 0.8–1.4 m combined with an H/W ratio close to 3:1 showed relatively favorable airflow-temperature performance in terms of shading continuity, moderated airflow, and reduced summer thermal exposure. The findings suggest that thermal comfort in compact hot–humid vernacular settlements depends on radiant-load reduction, moderated ventilation, and thermal buffering rather than on ventilation enhancement alone. Beyond the case-specific evidence, this study contributes a sequence-based, locally calibratable approach for preliminary retrofit appraisal in comparable compact hot–humid vernacular settlements. Full article

We investigate static, spherically symmetric (SS) spacetimes in covariant teleparallel $F\left(T\right)$ gravity in the presence of electromagnetic sources. Starting from the coframe/spin-connection (CSC) pair formalism, we derive the field equations and associated conservation laws, which constrain admissible electromagnetic configurations and reconstructed teleparallel sectors. A general reconstruction procedure is established, allowing the systematic construction of nonlinear teleparallel $F\left(T\right)$ models for arbitrary coframe ansätze. Focusing on power-law (PL) configurations, we obtain several classes of exact solutions, including constant-radius, black-hole-like (BH-like), and wormhole-like (WH-like) branches, and analyze their horizon structures, torsion singularities, and stability properties. The inclusion of electromagnetic sources leads to new charged solutions that generalize Reissner–Nordström (RN) spacetimes and reveal modified near-horizon and asymptotic behaviors. The results are further organized within an invariant classification framework, highlighting the role of torsion in shaping the solution space. Overall, this work provides a unified and covariant approach to the construction and interpretation of physically relevant compact-object, effective cosmological, and regularized strong-field sectors in nonlinear teleparallel gravity, with potential implications for strong-field tests beyond General Relativity (GR). Full article

Electron emission from hydrogen atoms induced by antiproton impact at intermediate energies is investigated using the one-centre Basis Generator Method within a semi-classical impact-parameter framework. The formulation employs a single-centre expansion of the time-dependent Schrödinger equation with a pseudostate basis consisting of hydrogenic orbitals acted upon by powers of a Yukawa-regularized potential, providing a compact and effective representation of the electronic continuum. Ionization probabilities are obtained by projecting the time-evolved wavefunction onto Coulomb continuum states, from which energy-differential cross sections (EDCS) are extracted. Exponential piecewise functions are constructed to interpolate between the pseudostate eigenenergies, yielding smooth EDCS profiles for each partial wave. The total EDCS, reconstructed by summing over all partial-wave contributions, exhibits good agreement with results from other pseudostate-based approaches. Full article

Motor imagery (MI)-based brain–computer interfaces (BCIs) have led to great interest as a result of their potential use in neurorehabilitation, assistive robotics, and human–computer interaction. However, decoding electroencephalographic (EEG) signals with high accuracy continues to be a difficult task due to the weak signal-to-noise ratio, differences among subjects, and the complicated temporal–spectral–spatial neural dynamics. Deep learning methods recently developed, such as convolutional neural networks, recurrent architectures, graph neural networks, and adversarial transfer learning, have enhanced MI decoding performance, yet many models are still concentrating on a single representation domain or they need costly adaptation phases in terms of computation. To tackle these shortcomings, we present NeuroCrossNet, a unified tri-modal deep learning model that is able to learn the temporal, spectral, and spatial EEG features jointly for robust and calibration-free MI decoding. The suggested network combines a Temporal HyperMixer Block for capturing long-range temporal dependencies, a wavelet transformer for learning localized time–frequency representation, and a Graph Attention Network for EEG topology-aware spatial reasoning. Additionally, a Dynamic Residual Attention Gate (DRAG) has been developed to adaptively merge heterogeneous feature streams, and a compact subject-aware normalization (SAN) method enhances cross-subject generalization without the use of labeled target-domain calibration data. Our proposed model was tested following the rigorous leave-one-subject-out (LOSO) approach on BCI Competition IV-2a and High-Gamma datasets. NeuroCrossNet reached a classification accuracy of 91.30%, surpassing several strong benchmark methods, including CNN-LSTM, EEGNet, DeepConvNet, spectral CNN, and graph-based EEG decoding frameworks. Furthermore, a large number of ablation studies reveal that the integration of temporally, spectrally, and spatially complementary representations considerably boosts robustness and inter-subject consistency. Full article

An optimum In_0.2Cd_0.8S composition was synthesized with graphene to enhance photocatalytic performance. Graphene incorporation altered the morphology from compact grains to a loosely aggregated structure without affecting the crystal phase, as confirmed by XRD. XPS analysis indicated surface-level interaction between graphene and the In–Cd–S matrix, rather than lattice integration. Mott–Schottky and Kubelka–Munk analyses revealed n-type semiconducting behavior and a slight band gap increase from 2.46 to 2.51 eV upon graphene blending. UV–Vis and IPCE measurements showed enhanced light absorption, with IPCE values of 9.33% and 5.01% at 380 nm and 480 nm, respectively. The 3.85 wt% graphene-modified photocatalyst achieved a hydrogen evolution rate of 4.97 μmolh⁻¹cm⁻², more than triple that of pristine In_0.2Cd_0.8S. These enhancements are attributed to improved charge transport and interfacial activity provided by the graphene. Full article

Falls are a major contributor to morbidity and mortality among older adults, and timely fall detection can help reduce the severity of fall-related outcomes. Wearable inertial measurement unit (IMU) sensors offer a promising solution for fall detection; however, many existing approaches rely on multiple sensing locations and computationally intensive models, which can limit their practicality for resource-constrained wearable devices. This study proposes a knowledge distillation framework for efficient wrist-based fall detection using the publicly available University of Málaga fall detection dataset (UMAFall), a benchmark dataset for human activity recognition and fall detection. Although UMAFall was not collected from older adults, it provides a useful public benchmark for evaluating IMU-based fall detection methods. Knowledge distillation was implemented using a teacher–student framework, in which a high-capacity teacher model trained with IMU data from four body locations (waist, wrist, ankle, and chest) provided soft targets for guiding a compact wrist-only CNN student model. In a held-out test evaluation using Subjects 2 and 5, the teacher model achieved 97.6% accuracy and an F1 score of 96.7%, with approximately 1.3 million trainable parameters. The independently trained wrist-based CNN achieved 90.2% accuracy and an F1 score of 87.1%. After applying knowledge distillation, the student model improved to 95.1% accuracy and an F1 score of 93.3% while maintaining the same lightweight architecture. A supplementary leave-one-subject-out analysis showed slightly higher and more stable AUC for KD-CNN than the independently trained CNN (0.96 ± 0.03 vs. 0.94 ± 0.07). These findings suggest that knowledge distillation can improve wrist-only fall detection in this feasibility evaluation, but further validation using older adults and real-world smartwatch data is needed. Full article

Rising global temperatures, increasing frequency and intensity of extreme climatic events, with associated growth of agricultural land use and urban expansion, represent critical drivers of biodiversity loss. Within this framework, urban areas are particularly vulnerable due to environmental stressors such as the heat-island phenomenon, soil sealing and depletion, and the accumulation of heavy metals and other pollutants. Recent sustainability-oriented urban policies recognize the strategic role of green infrastructures in mitigating these impacts by delivering essential ecosystem services, including phytoremediation. Here, the focus on herbaceous plants allows the selection of species with short life cycles and high colonization rates in marginal or disturbed urban habitats (e.g., roadside verges, compacted soils, and limited-volume planting areas). Therefore, the present review systematically examines herbaceous plant species with documented phytoremediation capabilities, focusing on Mediterranean native taxa evaluated under urban or peri-urban conditions. A total of 29 species met the selection criteria: taxonomically, Asteraceae represented the most frequent family (35%), followed by Fabaceae (21%), Brassicaceae, and Poaceae (each accounting for 10%). From a functional-trait perspective, hemicryptophytes dominated the dataset (66%), followed by therophytes (31%). Of the selected taxa, 55% primarily exhibited phytoextraction, 14% showed phytostabilization, and 31% demonstrated dual functionality, through combined extraction and stabilization pathways. These traits, combined with ecological adaptability to Mediterranean climatic regimes, support their application in Mediterranean urban environments. Full article

The formation of stereocomplex (SC) crystallites in poly(L-lactide) (PLLA)/poly(D-lactide) (PDLA) blends has attracted significant attention due to its potential to enhance the performance of biodegradable polymer films. In this study, the effect of solvent evaporation kinetics on the crystallization behavior, microstructure, and functional properties of PLLA/PDLA blend films was systematically investigated. Films with various blend ratios were prepared under open-lid (fast evaporation) and closed-lid (slow evaporation) conditions. Differential scanning calorimetry (DSC), wide-angle X-ray diffraction (WAXD), and small-angle X-ray scattering (SAXS) analyses revealed that slow solvent evaporation significantly promotes stereocomplex formation, particularly at the equimolar (50:50) composition, resulting in a higher degree of crystallinity and a more compact structure compared to fast evaporation conditions. These structural changes were directly correlated with improved functional properties. The optimized PLLA/PDLA (50:50) films exhibited a substantial reduction in water vapor permeability from 22.7 to 3.11 g·mm/m²·day·kPa (~86% decrease) and a marked decrease in microbial growth, as evidenced by reduced total plate count (TPC) values compared to neat polymers. The enhanced barrier performance and reduced microbial proliferation were attributed to the reduced free volume and increased tortuosity associated with densely packed stereocomplex crystallites, as supported by DSC and WAXD results. These findings demonstrate the importance of solvent evaporation kinetics in tailoring structure–property relationships to control stereocomplex formation and multiscale structural organization, providing a practical strategy for biodegradable packaging films. Full article

This study aims to develop CoCrMo/xCu composites through liquid phase sintering. The primary focus is on investigating how the addition of copper influences sintering kinetics, microstructure, and mechanical properties. The copper volume fraction ranged from 10 to 25 wt.% relative to CoCrMo. Sintering was conducted at 1150 °C under an argon atmosphere. Characterization methods included scanning electron microscopy, computed microtomography, and X-ray diffraction analysis. It was observed that molten copper, which forms upon reaching its melting temperature, can fill the interparticle spaces left by CoCrMo particles in the green compacts. During sintering, densification is further enhanced by the dissolution of CoCrMo, resulting in the formation of intermetallic phases enriched in Cr and Mo, as well as a ternary Co-Cr-Cu compound. Both densification and intermetallic formation contribute to increased microhardness as Cu content rises. It is concluded that the CoCrMo/25Cu composite exhibits the best mechanical and corrosion properties because its densification was improved by the Cu liquid. Full article

The growing diversity of wireless standards and complex real-world channel effects render automatic modulation classification (AMC) increasingly challenging for spectrum monitoring and edge intelligence. However, most competitive deep-learning-based AMC networks still require ${10}^5$–${10}^6$ parameters, exceeding the memory available on resource-constrained edge platforms. We propose LGP-Net, a lightweight gated-fusion network that pairs a physics-informed expert branch with a compact temporal encoder built from depthwise separable convolution (DSConv), squeeze-and-excitation (SE) attention, and a single-layer gated recurrent unit (GRU). Specifically, unlike other dual-branch structures that directly concatenate the outputs of both pathways, this work designs a lightweight gating unit that requires no external signal-to-noise ratio (SNR) labels and adaptively reweights the two pathways according to signal-quality degradation. With fewer than 40 K parameters, a peak activation footprint of 26.00 KB and an amortised inference latency of 9.7 $\mathsf{\mu }$s per sample under GPU acceleration, LGP-Net attains 65.00% overall accuracy on RadioML 2016.10B (91.48% at 0 dB) and 62.76% on RadioML 2016.10A, placing it in a competitive accuracy–efficiency regime relative to architectures consuming 5× to 500× more parameters. These characteristics support deployment-oriented feasibility under memory-constrained edge settings and high-throughput spectrum-monitoring pipelines. Full article

Off-road semantic segmentation is challenging due to irregular terrain, vegetation clutter, class-level similarity, and ambiguous boundary annotations. Existing decoder designs often rely on compact bottlenecks that oversmooth fine structures or repeated multi-scale fusion that can amplify annotation noise and increase computational cost. To address these limitations, we propose a Cross-Scale Decoder for robust off-road semantic segmentation. The proposed decoder first stabilizes semantic representations through Global–Local Token Refinement (GLTR) on a compact bottleneck lattice. It then selectively incorporates fine-scale structural cues using Boundary-Guided Correction (BGC) and Gated Cross-Scale Interaction (GCS), avoiding dense and repeated feature fusion. In addition, uncertainty-guided class-aware point refinement focuses computation on ambiguous and low-confidence regions. Experiments on standard off-road benchmarks demonstrate that the proposed method improves segmentation accuracy and boundary consistency over existing approaches while maintaining practical inference efficiency. Full article

Drug-induced QT interval prolongation is a critical safety concern in drug development, yet accurate and mechanistically interpretable prediction from chemical structure remains challenging due to the limited substructural resolution of existing approaches. Here, we present a motif-level graph learning framework for interpretable QT risk prediction. In this framework, molecules are decomposed into chemically meaningful motifs, enabling representation at an intermediate structural scale between atoms and predefined structural alerts. Motif features are encoded using a pre-trained chemical language model, and inter-motif relationships are modeled via attention-based graph learning with cross-scale integration. The model is trained and evaluated on two clinically grounded datasets derived from regulatory drug labeling (DIQTA) and real-world pharmacovigilance data (FAERS), achieving strong and consistent predictive performance with robust generalization across data sources. Importantly, motif-level attention reveals that QT liability is associated with the cooperative organization of compact cationic centers and heteroatom-rich, conformationally adaptable scaffolds, rather than isolated functional groups. These patterns are consistent with known determinants of human ether-à-go-go-related (hERG) channel blockade while providing a more structured and chemically specific interpretation beyond conventional structural alerts. Overall, this work establishes a generalizable and interpretable framework for QT risk prediction and highlights motif-level graph learning as an effective strategy for structure-based modeling of adverse drug reactions. Full article