Search Results (4,432)

Multi-task learning (MTL) is essential for satellite internet systems requiring simultaneous optimization of beam management, interference mitigation, resource allocation, and traffic prediction. However, existing evaluation methods rely predominantly on external performance metrics, neglecting internal dynamics governing task interactions. We propose TDS-Mamba (Task-Aware Decoupled State-Space Model), integrating selective state-space models with task-specific modulation for satellite networks. Our contributions include: (1) Task-Aware Decoupled S6 (TA-DS6) with hypernetwork-generated task-conditioned projection matrices; (2) Shared–Private State Decomposition disentangling cross-task representations from task-specific features; (3) Value-at-Risk (VaR) Gating for risk-sensitive optimization under varying orbital conditions; and (4) an internal diagnostic framework with Task-Specific Entropy and Interference Coefficient metrics. Experiments on LEO satellite constellation benchmarks show consistent improvements over the selected baselines and provide enhanced interpretability of multi-task dynamics via internal diagnostics. Full article

Large language models (LLMs) have shown strong performance in open-domain dialogue, yet they continue to struggle with long-term multi-session conversations (MSC), particularly in non-English languages such as Korean. In this work, we present a comprehensive empirical study on enhancing Korean MSC capabilities of LLMs through dataset construction, memory modeling, and parameter-efficient fine-tuning. We introduce an extended Korean MSC dataset that explicitly distinguishes between persona memory (long-term user attributes) and episode memory (short-term, event-driven information), enabling more effective memory management across sessions. Using this dataset, we evaluate LLM performance on three core MSC tasks: session summarization, memory update, and response generation. Our experiments reveal that Korean MSC is intrinsically more challenging than English MSC and that memory update and response generation require substantial reasoning ability. To address these challenges, we compare LoRA, DPO, MoE, CPT, Layer Tuning, and neuron-level tuning methods. Results consistently show that neuron tuning, guided by a novel language-specific neuron identification method based on activation scores and entropy, achieves superior performance and robustness, particularly in continual learning settings. Overall, our findings highlight neuron-level adaptation as an effective and interpretable approach for improving long-term conversational ability in low-resource languages. Full article

Noise contamination is a common challenge in the analysis of time series data, where stochastic perturbations can obscure deterministic dynamics and complicate the interpretation of signals from chaotic and physiological systems. Reliable identification of noise regimes and their intensity is therefore essential for robust analysis of dynamical and biomedical signals, where incorrect attribution of stochastic perturbations can lead to misleading interpretations of system behavior. For this reason, the present study examines the role of complexity-based descriptors for identifying stochastic perturbations in time series and analyzes how these metrics respond to different noise regimes across heterogeneous dynamical systems. A supervised learning approach based on complexity descriptors was developed to analyze controlled perturbations in multiple signal types. Gaussian, pink, and low-frequency noise disturbances were injected at predefined intensity levels into the Rössler and Lorenz chaotic systems, the Hénon map, and synthetic electrocardiogram signals, while AR(1) processes were used for validation on inherently stochastic signals. From these systems, eighteen entropy-based, fractal, statistical, and singular value decomposition-based complexity metrics were extracted from either raw signals or reconstructed phase spaces. These features were used to perform three classification tasks that capture different aspects of noise characterization, including detecting the presence of noise, identifying the perturbation type, and discriminating between different noise intensities. In addition to predictive modeling, the study evaluates the complexity profiles and feature relevance of the metrics under varying perturbation regimes. The results show that no single complexity metric consistently discriminates noise regimes across all systems. Instead, system-specific relevance patterns emerge. Under given experimental constraints (data partitioning, machine learning algorithm, etc.), Approximate Entropy provides the strongest discrimination for the Lorenz system and the Hénon map, the Coefficient of Variation, Sample and Permutation Entropy dominate classification for ECG signals, and the Condition Number and Variance of first derivative together with Fisher Information are most informative for the Rössler system. Across all datasets, the proposed framework achieves an average accuracy of 99% for noise presence detection, 98.4% for noise type classification, and 98.5% for noise intensity classification. These findings demonstrate that complexity metrics capture structural and statistical signatures of stochastic perturbations across a diverse set of dynamic systems. Full article

Five-axis Computerized Numerical Control (CNC) machine tools play a pivotal role in the precision manufacturing of aeroengine turbine blades, where ultra-high reliability and accuracy are essential. Failure Mode, Effects and Criticality Analysis (FMECA) has been widely applied in the reliability assessment of such advanced machining systems due to its systematic evaluation of potential failure modes. However, traditional FMECA approaches often overlook the ambiguity of human cognition and the interdependence among expert evaluations, limiting their effectiveness in complex aerospace manufacturing environments. To address these issues, this paper proposes a novel FMECA framework based on generalized intuitionistic linguistic theory. A new Generalized Intuitionistic Linguistic Weighted Geometric Average (GILWGA) operator is introduced to couple multi-source expert information and quantify the fuzziness inherent in subjective assessments. Additionally, an intuitionistic linguistic entropy-based weighting scheme is developed to dynamically evaluate key risk factors, including severity, occurrence, detectability, and controllability. The proposed framework is applied to a case study involving the spindle system of a five-axis CNC machine tool used in aeroengine blade production. The results demonstrate that the proposed method offers more robust and consistent failure mode prioritization, providing effective decision support for reliability-centered maintenance in aerospace equipment manufacturing. Full article

Type 1 diabetes (T1D) is an autoimmune disease with a strong genetic component (~70% heritability). Early identification of individuals at risk is crucial for early intervention or risk assessment. Although polygenic risk scores (PRS) have shown promise in risk assessment, most current approaches remain constrained by linear assumptions and limited generalizability. We aimed to develop a neural network-driven classifier using T1D-associated single nucleotide polymorphisms (SNPs). In addition, we explored the inclusion of an entropy-derived feature as a complementary variable, representing the degree of genetic variability within an individual’s genotype profile across the 67 T1D-associated SNPs, to evaluate its potential additive contribution to the model performance. We analyzed genotype data from 11,909 individuals in the UK BioBank (546 T1D cases and 11,363 controls). Sixty-seven well-known SNPs associated with T1D were utilized as inputs to the model, using two distinct allele-encoding strategies. A feed-forward neural network was evaluated under varying case–control ratios through five-fold cross-validation. Performance was assessed using the area under the receiver operating characteristic curve (AUC) on a held-out test set and on an external European cohort as a validation cohort. Across five-fold cross-validation, the best configuration achieved a median AUC of 0.903. On the held-out UK Biobank test set, the model generalized well, with an AUC of 0.8889 (95% CI: 0.8516–0.9262). A probability-based risk framework, constructed using five risk groups (“very low”, “low”, “intermediate”, “high”, and “very high” risk), yielded a negative predictive value (NPV) of 98.9% for the “very low” risk group and a Positive Predicted Value (PPV) of 61.9% with a specificity of 97.3% for the “very high” risk group, assuming a 10% T1D prevalence. External validation in the German Diabetes Study reproduced clear case–control separation; for individuals with recent onset diabetes and glutamic acid decarboxylase antibodies (GADA+) vs. controls, specificity reached 91.9% in the “high” risk group (PPV of 94.3%) and 97.6% in the “very high” risk group (PPV of 95.7%). The proposed neural network reliably predicts T1D genetic risk using a compact SNP panel of 67 SNPs and maintains accuracy in both internal and external European cohorts. Its probabilistic output enables clinically interpretable risk thresholds, while entropy features contributed modestly to performance. These results demonstrate that a neural network-based approach achieves discriminative performance that is comparable to established T1D genetic risk models, while offering flexible probability-based risk stratification and architectural extensibility for future integration of additional features. Full article

To address the limited expressiveness in current speech synthesis caused by coarse-grained prosody modeling and simplistic feature fusion strategies, a joint prosody modeling framework and a nonlinear fusion method named KAFusion are proposed, based on the Kolmogorov–Arnold (KA) representation theorem. The joint modeling integrates pitch and energy as prosodic priors with text encodings to jointly guide duration prediction, enabling explicit control over speech rate and tone. During feature fusion, KAFusion facilitates nonlinear interactions among features through its nested inner and outer functions. Information entropy serves as the quantitative metric, and both theoretical and experimental results demonstrate the fusion module’s efficacy in suppressing redundancy while preserving task-critical content. Evaluations on the AISHELL3 dataset show a 5.8% improvement in MOS over the baseline. Ablation studies further validate the effectiveness of the proposed components, where KAFusion achieves an output entropy of 3.47, which is 18.4% higher than that of linear fusion (2.93) and indicates richer information content. Full article

In this work, Fe, Co, Ni, Cu, and Mo powders were used as starting materials to prepare high-entropy alloy (HEA) thin films by a coating and vacuum sintering process. Using the HEA thin film as the substrate, selenium was subsequently deposited by chemical vapor deposition (CVD) to obtain high-entropy alloy selenide thin films (HEASe). The phase structure, surface chemical states, morphology, and elemental distribution of the porous films were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), and energy-dispersive X-ray spectroscopy (EDS). The electrocatalytic hydrogen evolution performance of the electrodes was evaluated using a three-electrode configuration in 0.5 M H₂SO₄, 1 M KOH, 1 M KOH + 0.5 M NaCl, and 1 M KOH + 0.5 M Na₂S solutions. The results indicate that the HEA selenide thin-film electrodes exhibit favorable electrocatalytic behavior in all four electrolytes. Among them, HEASe-450 shows the best overall performance. In 0.5 M H₂SO₄, it requires an overpotential of only 57.6 mV to reach a current density of 10 mA cm⁻², with a Tafel slope of 146.96 mV dec⁻¹. In 1 M KOH, the overpotential at 10 mA cm⁻² is 50.1 mV, and the corresponding Tafel slope is 142 mV dec⁻¹. In 1 M KOH + 0.5 M NaCl, the overpotential is 52.7 mV with a Tafel slope of 122.72 mV dec⁻¹. In 1 M KOH + 0.5 M Na₂S, an overpotential of 85 mV is required, and the Tafel slope increases to 236 mV dec⁻¹. Full article

Due to the rapid process of urbanization and the threat of environmental hazards, the need to enhance the intelligence and resilience of urban infrastructure has emerged as a pre-eminent demand of sustainable urban development. This paper evaluates the smart-resilience of urban infrastructure in Kunming by creating a well-developed evaluation framework with reference to the DPSIR (Driving Force–Pressure–State–Impact–Response) model and using the Entropy Weight TOPSIS technique to measure infrastructure performance during the years 2020–2024. The study fills an existing gap in the literature regarding the integration of intelligence and resilience evaluation, as well as the dynamic obstacle diagnosis based on causal logic. It provides a transferable analytical framework and empirical evidence for the “smart-resilience” development of similar cities. The findings suggest that there is steady progress in infrastructure smart-resilience in Kunming, whereby the composite index grew from 0.330 to 0.597, which is equivalent to an average growth rate of about 16.0 per annum. In spite of this favorable tendency, there are a number of structural issues that remain unsolved. The driving force dimension is unstable with regard to long-term mechanisms of investment, and the responding dimension is lagging behind, indicating weaknesses in the governance capacity and inter-departmental coordination. Moreover, extreme weather events have become the major threat to infrastructure systems in the city, superseding traditional social and operational risks; consequently, the city has changed its risk profile. Obstacle factor analysis shows that state and response dimensions make up almost 60% of the total constraint level, which shows the significance of enhancing the effectiveness of management. The research findings are based on the proposal of specific policy actions, such as the creation of special infrastructure resilience funds, the enhancement of mechanisms relating to cross-departmental emergency responses, the implementation of risk-based engineering standards, and the creation of an integrated infrastructure data platform to facilitate efficient, resilient, and sustainable urban governance. Full article

The aim of this study was to investigate which physicochemical and structural properties of cationic peptides P1–P6 may determine their selective anticancer activity against melanoma cells and their interactions with tumor cell membranes. An integrated approach was applied, including characterization in solution (osmotic pressure, NaCl stability, surface tension); cytotoxicity evaluation against Me45, B16F10, and HaCaT cells; analysis of interactions with phosphatidylglycerol (POPG) model membranes using isothermal titration calorimetry and steady-state fluorescence spectroscopy; membrane permeability assays; and F-actin staining. Anticancer activity depended on positively charged residues, hydrophobic amino acids, and sequence arrangement. Tryptophan-rich peptides P2 and P5 exhibited strong membrane interactions and high efficacy after 72 h. Highly hydrophobic P4, containing long C₁₂ chains with a relatively low net charge, caused nonselective lysis. P3 showed reduced activity due to insufficient amphipathicity, whereas P6, with excessive WWW and KKKK motifs, exhibited weak or nonselective effects. Thermodynamic and fluorescence analyses indicated that P2 and P5 initially bind POPG membranes via entropy-driven electrostatic interactions, followed by hydrophobic insertion of tryptophan residues, evidenced by increased fluorescence intensity and a blue shift of the emission maximum. P2, P4, and P5 induced actin cytoskeleton reorganization and increased membrane permeability, emphasizing the role of balanced amphipathicity and charge–hydrophobicity in designing selective anticancer peptides. Full article

High-quality agricultural development (HQAD) in China is essential to achieving Chinese-style modernization, which represents a uniquely Chinese path to modernization characterized by coordinated development across economic, political, cultural, social, and ecological dimensions. Against the backdrop of accelerating digitalization in global value chains (GVCs), exploring how it influences China’s HQAD carries significant theoretical value and policy implications. This study, for the first time, integrates GVC digitalization and HQAD into a unified analytical framework. Utilizing panel data from 30 Chinese provinces from 2009 to 2020, it empirically examines the relationship between them and the underlying mechanisms. GVC digitalization is measured with the interaction term between provincial digital GVC participation and global digitalization level, while HQAD is comprehensively assessed using a multi-dimensional evaluation indicator system constructed based on the new development philosophy, employing the entropy weight TOPSIS method. The findings reveal that GVC digitalization significantly promotes HQAD in China. For every one-standard-deviation increase in the degree of digitalization, the level of HQAD increases by an average of approximately 0.02 percentage points. Mechanism analysis further identifies industrial structure upgrading and rural integration of primary, secondary, and tertiary industries as two crucial transmission pathways. Heterogeneity analysis indicates that this promoting effect is more pronounced in major grain-marketing regions, provinces with better digital infrastructure, and those with higher levels of human capital. This research provides new empirical evidence for understanding agricultural transformation in the digital era and offers policy insights for leveraging GVC digitalization to advance HQAD. Full article

Background: Endovascular aneurysm repair (EVAR) requires lifelong imaging surveillance because endoleaks, aneurysm sac expansion, and severe adverse events occur in up to one-third of the patients. Conventional follow-up based on sac diameter and visual assessment may fail to detect early microstructural changes that precede clinical deterioration. Methods: This narrative review summarizes the current evidence on texture-based radiomics and artificial intelligence (AI) applied to computed tomography (CT) and CT angiography (CTA) for post-EVAR outcome prediction and surveillance. Original studies evaluating radiomic features and AI-based models for endoleak detection, aneurysm sac behavior, and EVAR-related adverse events were included and qualitatively synthesized. Results: Ten studies were included. Radiomic features describing texture heterogeneity, gray-level nonuniformity, entropy, and spatial complexity were extracted from the aneurysm sac, intraluminal thrombus, and perivascular adipose tissue. Machine learning and deep learning models achieved good to excellent performance, with reported AUC values ranging from 0.78 to 0.95 for predicting endoleaks, sac expansion, and severe adverse events. Texture-based radiomics consistently outperformed morphology-only assessments and showed complementary value to deep learning, including applications on non-contrast CT. Conclusions: CT texture radiomics combined with AI represents an emerging research approach with potential relevance for post-EVAR surveillance, although current evidence remains limited. By capturing tissue heterogeneity beyond conventional morphology, radiomics may enable the earlier detection of complications and support risk-adapted follow-up. However, the heterogeneity of methods limited external validation, and reproducibility issues remain major barriers to clinical translation. Full article

Effective water resource management requires understanding the causal mechanisms driving water quality dynamics, yet extracting actionable insights from complex multivariate monitoring data remains a persistent challenge. This study presents CEcBaN (CCM-ECCM-Bayesian Networks), a decision-support tool that integrates Convergent Cross Mapping (CCM) for detecting dynamical coupling, Extended CCM (ECCM) for identifying temporal lags and causal directionality, and Bayesian network (BN) modeling for probabilistic scenario-based inference. The tool was designed to enable managers and researchers without programming expertise to reconstruct causal networks from routine monitoring data, distinguish direct from indirect effects, and evaluate intervention scenarios. CEcBaN was validated using four synthetic datasets with known causal structures, achieving superior specificity (0.83) and edge count accuracy (25% error) compared to Transfer Entropy (0.47 specificity, 139% error), Granger causality (0.82, 39% error), and the PC algorithm (0.83, 46% error). Application to Lake Kinneret, Israel, demonstrated the tool’s utility across three water quality challenges: (1) nitrogen cycling, where the nitrification pathway was reconstructed and seasonal stratification was identified as a key modulator (accuracy 0.931); (2) thermal dynamics, where a transition from atmosphere-driven to internally regulated heat transfer during stratification was revealed (2.1-fold increase in coupling strength); and (3) cyanobacterial bloom prediction, where prior phytoplankton community composition provided a 4–6-week early warning window (accuracy 0.846). CEcBaN advances causal inference in water resource management by making these analytical methods accessible through an intuitive interface. Full article

Accurately assessing dynamic water resource carrying capacity (WRCC) is essential and challenging, particularly in regions like the Gansu sections of the Yellow River Basin (GSYRB), a core water source protection zone in the arid northwest of China, due to its pressing challenge of balancing water resources for socioeconomic needs and ecological security. This study proposes a novel integrated computational assessment framework named SD-VIKOR to address the complexities arising from nonlinear interactions within the “water resources–socioeconomic–ecological environment” (W–S–E) system. The core of this framework is the tight coupling of a system dynamics (SD) simulation model with a VIKOR multi-criteria evaluation module, where indicator weights are objectively–subjectively determined via an Analytic Hierarchy Process (AHP)–entropy weight method. This integrated SD-VIKOR engine enables dynamic, scenario-based WRCC trajectory simulation. To move beyond simulation and enable mechanistic insight, the framework further incorporates a diagnostic suite: a Geodetector module quantifies dominant drivers and their interactions; an obstacle degree model pinpoints key limiting factors; and a coupling coordination degree model evaluates subsystem synergies. Together, they form a closed-loop “dynamic simulation → multi-criteria assessment → driving mechanism analysis and constraint diagnosis → subsystem coordination analysis” workflow. Applied to the GSYRB from 2012 to 2030 under five development scenarios, the framework demonstrated high efficacy. It successfully captured path-dependent WRCC evolution, revealing that the ecological-priority scenario (B2), which shifts system drivers from economic-scale expansion to resource-efficiency and environmental governance, yielded optimal WRCC and the highest system coordination. In contrast, business-as-usual and single-minded economic expansion scenarios underperformed. Six key obstacle factors were quantitatively identified, linking WRCC constraints to natural endowments, economic patterns, and domestic demand. The results reveal pronounced spatial–temporal heterogeneity in WRCC across the GSYRB, with socioeconomic development, water resource use efficiency, and ecological conditions acting as the primary joint drivers of WRCC evolution. Critically, several key indicators are identified as persistent constraints on regional water sustainability. In contrast to conventional static evaluations, the integrated framework captures the complex dynamics and multi-subsystem interactions governing WRCC, offering a more robust diagnostic of resource–environment systems. These insights provide a transferable analytical basis for designing sustainable water management strategies in arid river basins. Full article

Deep learning models for brain tumor classification from magnetic resonance imaging (MRI) often achieve high in-dataset accuracy but exhibit substantial performance degradation when evaluated on unseen clinical data due to domain shift arising from variations in imaging protocols and intensity distributions. Existing approaches largely rely on architectural scaling or parameter-level regularization, which do not explicitly constrain the stability of learned feature representations. This manuscript proposes Feature Space Stability Regularization (FSSR), a lightweight and model-agnostic training framework that enforces consistency in latent feature representations under realistic, MRI-safe-intensity perturbations. FSSR introduces an auxiliary feature space loss that minimizes the ${\ell }_2$ distance between normalized embeddings extracted from the input MRI images and their intensity-perturbed counterparts, alongside standard cross-entropy supervision. This manuscript evaluated FSSR across three convolutional backbones, ResNet-18, ResNet-34, and DenseNet-121, trained exclusively on the Kaggle Brain MRI dataset. Feature space analysis demonstrates that FSSR consistently reduces mean feature deviation and variance across architectures, indicating more stable internal representations. Generalization is assessed via zero-shot evaluation on the fully unseen BRISC-2025 dataset without retraining or fine-tuning. On the source domain, the best-performing configuration achieves 97.71% accuracy and 97.55% macro-F1. Under domain shift, FSSR improves external accuracy by up to 8.20 percentage points and the macro-F1 by up to 12.50 percentage points, with DenseNet-121 achieving a 96.70% accuracy and 96.87% macro-F1 at a domain gap of only 0.94%. Confusion matrix analysis further reveals the reduced class confusion and more stable recall across challenging tumor categories, demonstrating that feature-level stability is a key factor for robust brain MRI classification under domain shift. Full article

Improving the living environment in rural areas is an important task and a key breakthrough point in implementing the rural revitalization strategy. It not only directly affects the vital interests and health protection of farmers, but is also an important measure to promote ecological civilization construction and achieve the development goal of a beautiful China. Taking environmental demonstration villages in Shaanxi Province as the research object, questionnaire data were obtained through field research and face-to-face interviews. This study constructs an evaluation index system covering five dimensions: village appearance, domestic sewage treatment, rural toilet renovation, domestic waste treatment, and construction and management mechanism. The entropy method is used to determine indicator weights, and fuzzy comprehensive evaluation is applied to measure the implementation effect. The research results indicate that the overall effect is between “average” and “good” (score 3.924), with domestic sewage treatment scoring highest and construction and management mechanism lowest. The study identifies key problems such as low farmer participation, insufficient funding sources, inadequate infrastructure maintenance, and weak environmental awareness. Based on these findings, countermeasures are proposed: enhancing farmers’ environmental awareness and participation; diversifying capital investment; improving infrastructure and establishing long-term management mechanisms; cultivating social capital; and strengthening the leading role of the government. This study provides empirical evidence and policy recommendations for improving rural environmental governance. Full article