Urban Heat Islands (UHIs) intensify thermal stress in fast-growing cities, with wetlands functioning as major cooling nodes. However, how this cooling spreads through urban space remains poorly understood, largely because existing models rely on simplified radial assumptions that ignore the influence of street-network form. This study proposes a conceptual framework that enables quantitative evaluation of how street-network configuration shapes the transmission of wetland-driven cooling within urban heat island conditions, addressing current gaps related to airflow directionality, three-dimensional turbulence, and the obstruction of cool-air movement by block structure, network permeability, and street pattern. The study integrates spatial, spectral, morphological, and microclimatic urban heat indicators with key wetland variables including area, hydrological function, and vegetation structure, drawing on methodologically aligned case studies. Street-network metrics derived from space syntax and network science such as integration, choice, and connectivity are incorporated as measurable parameters linked to airflow and heat dispersion. The framework organizes urban heat island factors as the primary index, wetland attributes as mediators, and street-network characteristics as modulators, enabling systematic examination of their interdependencies across diverse urban settings. The literature review shows that while wetland cooling can be quantified based on landscape properties, research has not adequately modelled how this cooling interacts within street networks. The framework developed here offers a transferable method for guiding urban design decisions, allowing planners to configure street networks that enhance cool-air penetration, strengthen blue–green infrastructure performance, and support UHI mitigation.

Predicting the Li-ion migration energy barrier in battery cathode materials via machine learning has attracted increasing attention. However, the capture of complex material features via the accurate definition of structural descriptors and cross-scale information integration for neural network modeling remains challenging. Benefiting from the reported divide-and-conquer strategy, we propose a convolutional neural network (CNN) model employing hybrid geometric-topological descriptors for the efficient prediction of Li-ion migration energy barriers, in which the geometric descriptors capture local Li-O polyhedra of initial and transition states, as well as the topological descriptors derived from persistent homology characterize structural connectivity and ring channels along the migration pathway. Compared with the recurrent neural network (RNN) and Fourier-feature network (FFN) models, the CNN model optimized via residual block structures and L2 regularization achieves a mean absolute error (MAE) of 0.0589 eV. The minimum Li-O distance during the Li-ion migration is identified as the most critical factor affecting the migration barrier, and the comparable importance scores of dmin, dstd, and H0 bars suggest a synergistic effect of multiple structural descriptors, highlighting the necessity of adopting hybrid geometric-topological descriptors. The present work provides an efficient and accurate approach for high-throughput screening of materials with rapid Li-ion diffusion.

Rapid organic acid mechanochemical extraction of alginate from Laminaria digitata.

DynaHash: An efficient blocking structure for streaming record linkage

The design and simulation of an inductive coupler with an integrated DC bias function are reported. A choke structure is implemented to reduce the electric-field on the ceramic window, lowering sparking risk. Particle-in-cell simulations reveal the power-dependent behavior of multipacting, showing electron cloud migration towards the ceramic at high power, which is correlated with increased thermal deposition in thermal analysis. To enable multipacting suppression via DC bias, a blocking structure featuring a Kapton-film and ceramic-copper composite insulator is introduced between the coupler's inner and outer conductors. This study provides a validated design approach for stabilizing high-power CW inductive couplers against multipacting, forming the basis for subsequent prototyping and conditioning.

Convolutional Neural Networks (CNNs) excel at extracting local features, but due to their restricted receptive fields, they often struggle to capture large-scale global context. Transformers leverage self-attention mechanisms to facilitate global interactions, yet the computational cost of standard self-attention scales quadratically with image resolution. To overcome these limitations, we propose MFDA-UNet, which adopts a hybrid architecture of convolution and linear attention for synergistic feature processing. To fully leverage their respective strengths, we design the Mamba-inspired Frequency-Decoupled Attention (MFDA) block. Through frequency decoupling, this block utilizes convolutions to process high-frequency local information, while employing linear attention to model the long-range dependencies of low-frequency global information. To enhance the feature representation capability of linear attention, we construct the Mamba-Enhanced Linear Attention (MELA) block. Inspired by MILA, this block injects Positional Encoding to substitute the forget gate functionality of Mamba and integrates the Mamba block structure into the linear attention mechanism. This design effectively strengthens representational power, accomplishing long-range dependency modeling with highly efficient linear complexity. Furthermore, we introduce the Gated Cross-Scale Attention (GCSA) module to optimize traditional skip connections. It aggregates features via cross-scale linear attention and incorporates Mamba’s high-performance gating mechanism for adaptive feature filtering, achieving precise feature fusion and selection. We conducted extensive experiments on four multi-modal benchmarks: ISIC 2017, ISIC 2018, Synapse, and ACDC. MFDA-UNet achieved improvements in the DSC by 0.44%, 0.15%, 0.53%, and 0.84% across the respective datasets compared to the second-best models. By capturing local and global multi-scale semantics with relatively low computational overhead, MFDA-UNet provides an efficient and robust solution for medical image segmentation.

This essay explores how Seoul became both a stage for urban planning and an actual site of construction in the late 1960s. The city government and its mayor, Kim Hyŏn-ok, avidly proposed numerous brand-new city designs and construction plans through various media, among which a miniature of 1980s Seoul at the Urban Planning Exhibition (1966) and three-dimensional designs of the Yŏŭido Development Plan were representative cases. Further, those construction projects were implemented as ongoing everyday events in association with a specific discourse, kŏnsŏl. This term was a key concept for urban planning then, indicating the whole process that not only embarked on and carried out those city plans but also changed the city's environment from vacant, canvas-like areas that were created by explosive destruction to multidimensional landscapes with structures of steel frames and concrete blocks, such as high-rise buildings, overpasses, expressways, and embankments that literally demonstrated a Seoul under construction.

Spectral fluctuation analysis within the random matrix theory (RMT) framework is a powerful probe of complexity in networked systems, yet its extension to multilayer architectures remains unresolved. In general multilayer networks, the adjacency matrix possesses a heterogeneous block structure with unequal variances across layers, causing eigenvalue spacing statistics to deviate from RMT predictions even when each individual layer is perfectly random. We demonstrate that this variance mismatch is the central obstacle to observing spectral universality in multilayer systems and introduce a general blockwise normalization scheme that restores the correct variance structure across all blocks. Using higher-order spacing ratios, we show that once properly normalized, multilayer networks exhibit universal spectral fluctuations consistent with RMT across a broad class of configurations, including purely intralayer, interlayer, and multiplex structures. Focusing on the bilayer case, we introduce a crossover model parametrized by the relative interlayer to intralayer coupling strength, which captures the continuous transition from two independent Gaussian orthogonal ensembles (GOEs) to a single GOE. We find that this crossover sharpens with increasing system size, suggesting that in the large-system limit, arbitrarily weak interlayer coupling may be sufficient to induce global spectral correlations. Applying the framework to empirical multilayer networks derived from protein-crystal structures, we demonstrate that structural coupling drives analogous spectral transitions, directly linking the emergence of universality to physically meaningful organization. These results establish spectral universality as a robust feature of multilayer networks and provide a quantitative framework for understanding how structure and coupling govern collective behavior in complex interconnected systems.

ABSTRACT We introduce a dynamic factor correlation model whose core methodological innovation is a variation‐free parametrization of dynamic factor loadings, inspired by the generalized Fisher transformation. The model accommodates time‐varying correlations, heterogeneous heavy tails, and dependent idiosyncratic shocks. Applied to a Small Universe of 12 assets and a Large Universe of 323 stocks, the factor structure induces a sparse idiosyncratic correlation matrix with dependencies concentrated within subindustries, enabling scalability to high dimensions under a sparse block structure. Both factor loadings and correlations vary substantially. Allowing for heterogeneous heavy tails via convolution‐ distributions yields sizable improvements relative to Gaussian and multivariate‐ benchmarks.

Slash pine (Pinus elliottii) possesses an exceptionally large, repeat-rich genome, and existing low-density SNP arrays provide limited marker coverage and resolution of linkage patterns. To increase marker density for population-based genetic analyses and improve the accuracy of genomic relationship matrix (GRM) estimation, we constructed a reference panel from about 10× whole-genome resequencing of 50 maternal parents and performed genome-wide imputation for 51K SNP-array genotypes of 715 half-sib progeny. Imputation accuracy at array loci was quantified using chromosome-local masking experiments, whereas reference-panel-expanded loci not represented on the array were evaluated for concordance and filtered via external validation using progeny resequencing data. Masking-based concordance remained stable at 95.5%, and after threshold-based filtering of expanded loci, we generated a high-density genotype matrix for all 715 individuals comprising 120,650,180 SNPs. Comparative local linkage disequilibrium (LD) heatmaps indicated more continuous LD signals and clearer block structures after densification; for the Chr4 10.22-10.33 Mb interval, the proportion of high-LD SNP pairs increased from 14.5% to 27.6%. The GRM derived from the densified dataset was highly consistent with the array-based GRM in off-diagonal elements (Pearson's r≈0.984). Distance-stratified analyses further showed higher concordance for GRMs constructed from imputed loci within 500 kb of array markers, with concordance decreasing progressively in more distant windows, suggesting limited incremental benefit from long-range imputed loci. Collectively, the reference panel-driven imputation, validation, and integration framework established here provides a high-density genotypic resource for genome-wide association studies and genomic selection in slash pine and other conifer species.

Graphical models serve as effective tools for visualizing conditional dependencies between variables. However, as the number of variables grows, interpretation becomes increasingly difficult, and estimation uncertainty increases due to the large number of parameters relative to the number of observations. To address these challenges, we introduce the Clusterpath estimator of the Gaussian Graphical Model (CGGM) that encourages variable clustering in the graphical model in a data-driven way. Through the use of an aggregation penalty, we group variables together, which in turn results in a block-structured precision matrix whose block structure remains preserved in the covariance matrix. The CGGM estimator is formulated as the solution to a convex optimization problem, making it easy to incorporate other popular penalization schemes which we illustrate through the combination of an aggregation and sparsity penalty. We present a computationally efficient implementation of the CGGM estimator by using a cyclic block coordinate descent algorithm. In simulations, we show that CGGM not only matches, but oftentimes outperforms other state-of-the-art methods for variable clustering in graphical models. We also demonstrate CGGM’s practical advantages and versatility on a diverse collection of empirical applications.

ABSTRACT This paper examines how deregulated planning, particularly Permitted Development Rights (PDR), has reshaped residential and urban morphologies in London through incremental adaptations. Using morphological and visual analysis across two contrasting London districts, the study documents how small-scale extensions – rear, side, and rooftop – accumulate over time to generate new hybrid typologies. While intended to remain invisible, these interventions produce a form of morphological informalisation within a formally regulated context. The findings reveal how soft densification transforms dwelling types, block structures, and visibility in the cityscape, highlighting the role of residents as co-producers of urban form.

Translating programmatic assessment for learning (PAL) across clerkship models: implementing PAL in block and longitudinal structures

Abstract We present a novel algorithm for quantization and subsequent hexahedral mesh generation from seamless volumetric maps. Quantization is the process of choosing integers that represent the numbers of hexahedral elements to be placed in each region of the volume, and transforming the seamless map into an integer‐grid map matching that choice, inducing a hexahedral mesh. Previous work computes such quantizations under the restriction of a fixed predetermined singularity graph. Our novel approach allows for implicit modification and, in particular, simplification of the map's singularity structure wherever that benefits the chosen objective, such as matching target hexahedron sizes as closely as possible. It comes with two novel ingredients: A feature‐focused distortion measure guiding the quantization, and constraints ensuring map injectivity and structure preservation of geometric and topological features, both without relying on a fixed singularity structure. We demonstrate the benefit of the added flexibility offered by this approach: it allows for the generation of hexahedral meshes that more accurately match a desired resolution globally, as well as of meshes exhibiting a simpler block structure.

We demonstrate, for the specific case of C3v, how the direct-product decomposition scheme for the treatment of symmetry in coupled-cluster (CC) calculations can be extended to non-Abelian point groups. We show that for the two-electron integrals and CC amplitudes, a block structure can be obtained by resolving the reducible products of two irreducible representations into their irreducible representations. To deal with the necessary re-sorts of the ordering of the two-electron integrals and amplitudes, spin adaptation, and the contractions (with M as the number of basis functions) of a CC calculation, we suggest a strategy that uses both the reduced and nonreduced representations of the corresponding quantities and switches back and forth between them. While the reduced representations are the ones used in the contractions, the other steps are better carried out in the nonreduced representation. Our pilot implementation of the CC singles and doubles method confirms in test calculations for NH3 and PH3 using different basis sets that significant savings (of more than 20 compared to treatments without symmetry and about 5 compared to treatments using Cs symmetry) are possible and these findings suggest that the exploitation of non-Abelian symmetry would render CC computations on large, highly symmetric molecules possible.

Cross-scale effects and driving mechanisms of spatial structure and functional coupling in coastal city leisure blocks: A case study of the Jiaodong urban agglomeration

The paper presents the results of Petrov’s 3D inversion applied to gravity data to determine the continuous distribution of rock density from the surface down to a depth of Z = 7250 m in the Nong Son - Da Nang area, supporting the delineation of potential zones associated with ore-forming processes. The inversion method was performed continuously on the residual gravity field using a two-dimensional “live-window” energy filter, with window sizes varying from 600÷8600 m. The reliability of the inversion results strongly depends on the accurate determination of residual gravity anomalies. Therefore, in this study, we apply a statistical-probabilistic approach to identify adaptive filter-window geometries that are consistent with the regional anomaly trend, thereby enhancing the accuracy of residual gravity anomaly separation. The results show that, at the southwestern ore-point locations, the subsurface rock density is heterogeneous, forming blocks with positive residual-density values ranging from 0.1÷0.35 g/cm³, corresponding to residual gravity anomalies of 1÷3.5 mGal. The residual density anomalies within local block structures at the ore-point sites extend to depths of approximately 4000 m. The density increases from about 2.25 g/cm³ at the surface to 3.05 g/cm³ at a depth of 7250m. In the ore-bearing zones, the density varies from 2.8 g/cm³ near the surface to 3.0 g/cm³ at depths of around 4000 m, forming a continuous band extending upward from depth toward near-surface levels. The interpretation results indicate potential mineralized zones associated with the upward migration and near-surface accumulation of magma-derived materials.

Blockchain technology continues to promise transformative impact across domains such as supply chains, finance, and the Internet of Things (IoT). However, the rapid growth and increasing heterogeneity of blockchain platforms have made architectural decision-making progressively more complex for software architects. This study extends and updates a previous Multivocal Literature Review (MLR) to systematically identify and characterize active blockchain networks across foundational protocol layers. We analyze key architectural dimensions including consensus mechanisms, decentralization and access control models, smart contract support, block and ledger structures, interoperability features, and architectural lineage. Drawing on both academic and gray literature, we characterize a total of 147 blockchain networks spanning Layers-0 through-2. Our findings reveal an ecosystem largely driven by industrial innovation, with limited consolidation in the formal academic literature. The resulting architectural mappings aim to support software architects in making informed, evidence-based decisions when integrating blockchain technologies into software-intensive systems.

Pattern-coupled Bayesian compressive sensing shows great potential in sound field reconstruction by leveraging structural sparsity, but its fixed coupling patterns for sparsity hyperparameters limit adaptability to non-uniform correlation distributions. To overcome this limitation, this paper proposes an enhanced method termed data-adaptive pattern-coupled Bayesian compressive sensing for high-accuracy sound field reconstruction. In this method, a hierarchical Gaussian-Gamma prior model is established based on the equivalent source method within the compressive sensing framework, achieving reconstruction by solving for the sparse coefficient vector of equivalent source strengths. A set of adaptive coupling parameters is introduced via a learnable transformation matrix, dynamically regulating the interrelationships between hyperparameters and thereby substantially enhancing the adaptability of the prior model. Furthermore, both the coupling parameters and hyperparameters are iteratively updated with a data-driven method, enabling adaptive mutual influence of sparsity patterns among elements within the sparse coefficient vector. This process promotes clustering of non-zero coefficients and concentration of zero-valued coefficients, inducing a physically meaningful block-sparse structure reflecting the spatial continuity of actual sound sources. By fully exploiting the intrinsic statistical correlations between elements of the sparse coefficient vector without requiring knowledge of the block structure, it achieves superior sound field reconstruction accuracy. Numerical simulations and experimental results demonstrate that the proposed method outperforms existing approaches in terms of reconstruction accuracy and noise robustness, thereby validating its effectiveness and superiority in sound field reconstruction.

ABSTRACT In federated learning for large‐scale edge nodes, the problem of malicious clients submitting anomalous parameters is becoming increasingly prominent. This seriously affects the accuracy and reliability of the model. For the problem of malicious client defense in a large number of clients with decentralized distribution, to reduce the detection time of malicious clients in federated learning and improve the accuracy of model training, a Blockchain‐based Grouped Federated Learning malicious client defense Scheme (BGFLS) is proposed. Specifically, to detect malicious clients quickly and accurately, a grouped federated learning architecture is proposed, which applies blockchain technology to each grouping. In addition, an algorithm is designed to detect anomalous parameters, and a block structure that supports backtracking of malicious clients is proposed. Theoretical analysis and experiments show that the BGFLS scheme has improved accuracy compared with the GeoMed scheme and Krum scheme, and its backtracking efficiency is better than that of traditional blockchain implementations. Therefore, the BGFLS scheme can quickly detect malicious clients and protect shared parameters. This study provides a practical and high‐performance solution for detecting malicious clients using federated learning in large‐scale edge computing environments, with excellent technical specifications and high operational efficiency, effectively optimizing the overall system performance and stability.