Abstract: This paper explores the fundamental nature of spacetime from the perspective of emergent causal structures, rooted in quantum mechanical principles. We propose a theoretical framework that treats spacetime not as a pre-existing background, but as a collective phenomenon arising from the interaction of quantum causal relations. By analyzing the constraints of causality and quantum coherence, we derive key implications for the emergence of classical spacetime geometry and the limits of local realism. Our results suggest a new way to bridge quantum mechanics and gravitational theory, providing a foundation for future studies in quantum gravity and spacetime physics.

Abstract: In this paper we study competition between AI providers when users cannot fully verify model quality. Many AI services are not well described as standard search goods, and they are not pure experience goods either. Price, interface quality, and latency are usually observable, but reliability, hallucination risk, and the downstream cost of error are often only imperfectly observable even after use. Building on the emerging view that algorithmic advice can exhibit credence-good features, we embed that insight in a static industrial-organization model of vertical quality differentiation, costly certification, and limited user comprehension. Two firms choose whether to offer a low-quality or high-quality model. High-quality AI reduces error risk, but it is slower and costlier. A high-quality firm can purchase credible certification or disclosure, yet only a subset of users can interpret it. We characterize the pure-strategy equilibrium set, show how low-quality pooling can arise even when superior technology exists, and identify a quality-trap region in which the unique market equilibrium is low-quality pooling although an allocation with one high-quality provider is welfare superior. We then analyze policy. Standardized certification works through the demand side by increasing the fraction of users who can reward quality; minimum quality standards work directly but bluntly; liability shifts firms' cost incentives and weakly shrinks the region in which a low-quality industry outcome can be sustained. Contrary to a common rhetorical move in AI policy debates, these instruments are not interchangeable. The model also clarifies that our framework is a certification model rather than a full Grossman-Milgrom unraveling game: the key distortion comes from limited user comprehension of costly, truthful quality communication. The results offer a tractable industrial-organization foundation for current debates over hallucinations, model evaluation, AI documentation, and governance.

Abstract: Physics-informed neural networks (PINNs) have emerged as powerful tools for solving partial differential equations, but their training remains challenging due to ill-conditioned loss landscapes. While adaptive methods like Adam dominate deep learning, they exhibit instability on stiff PDEs, and second-order methods are computationally prohibitive. We present EPANG-Gen, an optimizer that combines memory-efficient eigen-decomposition with lightweight Bayesian uncertainty quantification. EPANG-Gen introduces three ele- ments: (1) a randomized eigenspace estimator that approximates Hessian curvature with O(dk) memory (k ≪ d), (2) Bayesian R-LayerNorm for per-activation uncertainty estima-tion, and (3) adaptive rank selection (PASA) that dynamically adjusts to problem difficulty. We evaluate EPANG-Gen on four benchmark PDEs—Poisson 1D, Burgers’ equation, Darcy flow, and Helmholtz 2D—and on the Taylor-Green vortex at Re = 100, 000, a canonical 3D turbulence problem. All experiments were conducted under computational constraints (Kaggle, NVIDIA P100 GPU, limited epochs). Results show that EPANG-Gen achieves performance comparable to Adam on the toughest turbulent regime while eliminating the 25% catastrophic failure rate of ADOPT across 72 runs. Ablation studies confirm that eigen- preconditioning contributes to performance improvements of 11–35%. The built-in uncer- tainty estimates provide confidence metrics at negligible cost. This work represents an initial exploration of curvature-aware optimization for PINNs; further validation with larger com- pute resources is needed. Code is available at https://github.com/EPANG-Gen/EPANG-Gen.

Abstract: Radio signals carry information in three natural ways: by changing how strong the signal is (amplitude), how high or low its tone is (frequency), and how its timing shifts within the wave (phase). In most communication systems, engineers use only one of these features at a time. As a result, much of the signal’s potential to carry information remains unused. This paper explores a simple but powerful idea: using all three features of a radio wave simultaneously to transmit information on a single carrier signal. By combining amplitude, frequency, and phase modulation together, a single radio wave can carry far more information without requiring additional bandwidth.To explain and analyze this concept, the work introduces an intuitive geometric framework inspired by a four-dimensional shape called a \emph{tesseract}, often described as a “four-dimensional cube.” In this framework, three directions represent the three information channels—amplitude, frequency, and phase—while the fourth represents time. This geometric picture provides a clear way to visualize how the three channels coexist without interfering with each other.As a simple demonstration, the phrase “I Love You” is encoded by assigning each word to a different feature of the signal: “I” is carried by amplitude changes, “Love” by frequency variations, and “You” by phase shifts. Colourful waveform plots, three-dimensional visualizations, and a novel “tesseract slicing” illustration help make the four-dimensional behaviour easier to understand.The proposed framework has potential applications in satellite communication, future 5G/6G networks, radar systems, and signal-processing education. By using all three dimensions of a signal at once, this approach reveals previously unused communication capacity and shows how a single radio wave could deliver substantially more information without consuming extra spectrum.

Abstract: The rapid proliferation of unmanned aerial vehicles (UAVs) in energy-intensive applications (such as autonomous logistics, continuous surveillance, and mobile edge computing) has driven a critical need for highly reliable energy consumption models. However, selecting an appropriate modeling strategy remains an ad-hoc process; researchers must frequently navigate complex, undocumented trade-offs among required predictive accuracy, empirical data availability, and access to aerodynamic testing infrastructure. This study proposes a systematic, two-stage decision-making framework designed to standardize UAV energy model selection. In the first stage, a qualitative decision tree is inductively derived from a comprehensive corpus of recent literature (an 80% training split), explicitly mapping infrastructural and informational constraints to five distinct modeling regimes, ranging from novel white-box derivations to deep-learning black-box applications. This structural logic is subsequently validated against an independent 20% literature holdout set, achieving a 100% predictive match. In the second stage, the Analytic Hierarchy Process (AHP) is applied to quantitatively rank the feasible alternatives based on context-specific criteria: accuracy, interpretability, development cost, and customization adaptability. Crucially, this quantitative scoring introduces "fallback flexibility," allowing researchers to seamlessly pivot to mathematically adjacent alternative models when unforeseen experimental roadblocks occur. Embedded within an open-source Python graphical interface, this framework mitigates methodological ambiguity, prevents the over-allocation of research resources, and fosters greater reproducibility within the energy-aware UAV research community.

Abstract: Background: Variation in suspected serious adverse events (SAEs) linked to different batches of COVID-19 vaccines has been reported in several countries, including the Czech Republic, Denmark, Sweden, and the USA. However, SAE data come from spontaneous reporting systems and are subject to under-reporting and other biases. We examined all-cause mortality (ACM) data to explore the temporal relationship between COVID-19 vaccine type and batch, and death, up to three months after vaccination. Methods: We analyzed nationwide data from the Czech Republic on vaccine type and batch, along with the corresponding three-month ACM data. Cluster analysis was used to assess differences in ACM across vaccine types and batches. Cluster-specific mortality rates were adjusted for age and sex and compared, with a focus on the timing of batch administration. We also investigated the relationship between ACM and SAEs for the same batches. Results: During a 21-month period (December 2020 to September 2022), vaccine batches were grouped according to three-month ACM rates for the four products administered (Comirnaty, SPIKEVAX, Vaxzevria, and Jcovden). For Comirnaty, SPIKEVAX, and Vaxzevria, a clear temporal pattern appeared, with earlier batches showing significantly higher ACM rates, even after adjusting for age and sex. A strong correlation was found between batches clustered by mortality and those previously identified to cluster by reported SAEs for all products except Jcovden. Conclusions: Data from the Czech Republic reveal a clear link between the most recently administered COVID-19 vaccine batch and short-term ACM rates. For three of the four vaccines, earlier batches were associated with notably higher ACM. The similar pattern observed between batch-associated mortality and SAE rates supports the existence of batch-related safety signals that warrant further investigation using individual-level patient data.

Abstract: We deploy a spiking neural network (SNN)-equivalent intrusion detection system (IDS) on the STM32N6570-DK, a commodity ARM Cortex-M55 MCU with the Neural-ART NPU. Exploiting the approximate equivalence between single-timestep (T=1) SNN inference and INT8 quantized ANN inference, we compile a lightweight MLP classifier to the NPU without neuromorphic hardware. Evaluated on NSL-KDD (5-class) and UNSW-NB15 (10-class) with 10 random seeds, the ReLU model achieves 78.86±1.32% and 64.75±0.61% overall accuracy, respectively. INT8 accuracy stays within 1 percentage point of FP32 across all 24 tested calibration configurations, and layer-wise analysis shows 99.0% final prediction agreement between FP32 and INT8 models. On the NPU, the INT8 model infers in 0.46 ms on NSL-KDD and 0.29 ms on UNSW-NB15 (100% NPU execution), 2.7–4.2× faster than the same model on the Cortex-M55 CPU, while occupying 120.6–137.7 KB Flash and 0.5–1.25 KB RAM. Tree-based baselines (Random Forest, XGBoost) achieve higher overall accuracy on UNSW-NB15 but cannot be compiled for the NPU at all. To our knowledge, this is the first publicly documented IDS deployment on an ARM Cortex-M NPU and the first publicly documented empirical validation of T=1 SNN–ANN equivalence on commercial NPU silicon.

Abstract: Strong light interference severely degrades imaging system performance. This paper presents a novel Digital Micromirror Device (DMD)-based imaging system for robust strong light suppression and long-distance detection. Our design strategically places the DMD at the primary image plane, utilizing a large F-number objective for extended depth of field. The relay imaging system employs a tilted image plane in a near-symmetric configuration to effectively balance DMD-induced aberrations, simplifying alignment and achieving a compact, high-performance layout. The DMD's regional flipping capability enables precise, dynamic suppression of strong light. Experimental results from a fabricated prototype demonstrate superior imaging quality (MTF > 0.3 at 167.3 lp/mm) and exceptional suppression of intense laser interference, ensuring clear image acquisition in challenging lighting. This system offers an efficient solution for high-quality, long-range imaging in strong light environments.

Abstract: Insect attractant lures come in many formats, one of which utilizes tapered rubber sleeve stoppers, normally utilized to seal laboratory glassware openings. Their cup-shape top happens to be ideal to pipet a solution within this cavity, and, through permeation, load quantities of active ingredients. The expansion or swelling of the rubber facilitates the permeation of the active within its matrix, a role that dichloromethane performs well. Dichloromethane is also favored due to its volatility and broad chemical compatibility. However, this solvent is possibly on the verge of retirement, which would mean finding alternatives. It was found that several other common laboratory solvents could serve as replacement, and of those tested, tetrahydrofuran outperformed dichloromethane in terms of overall volume uptake and swelling. When loading the septum/sleeve with larger amounts of active, a full soaking methodology can disperse the active throughout the rubber sleeve as well as reduce labor requirement since batches can be processed compared to manually pipetting a solution to individual sleeves.

Abstract: In recent years we have seen Large Language Models (LLMs) demonstrating robust reasoning capabilities comparable to human performance. This makes them increasingly appealing for driver assistance, where adaptation to dynamic human context is essential. Yet, research in this area remains fragmented, often focusing on isolated applications, lacking utilization of LLM's full potential to deliver integrated, context-specific support and action. This survey synthesizes recent advancements in LLM-driven occupant monitoring systems, focusing on their capabilities for interpreting driver states and acting appropriately, enabling a new generation of intelligent driver assistance. We critically examine pioneering frameworks, benchmarks, and foundational datasets that employ techniques like reasoning chains, multimodality, and human-in-the-loop feedback to create personalized and safe driving experiences. We lay out the current trends, limitations, emerging patterns, in addition to a novel human-centered evaluation of the field, providing researchers with a roadmap towards transparent and trustworthy in-cabin systems, that bridge safety with driver experience.

Abstract: Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common chronic liver disease in children and is strongly associated with obesity and insulin resistance. In this study, we evaluated the clinical effects of GLP-1 RA therapy in a de-identified cohort of pediatric patients with MASLD and investigated potential molecular mechanisms using publicly available transcriptomic datasets from models of liver disease. Longitudinal FibroScan measurements from seven pediatric patients treated with GLP-1 RAs demonstrated significant reductions in controlled attenuation parameter scores, transient elastography scores and AST levels, indicating improvements in hepatic steatosis, liver stiffness and the liver inflammatory profile, respectively. To explore potential mechanisms underlying these observations, we analyzed transcriptomic datasets from methionine-choline deficient (MCD) and high-fat diet (HFD) mouse models of liver disease. A pattern-matching algorithm identified a core set of ten genes consistently upregulated in both models and downregulated following GLP-1 RA treatment in the HFD model. These genes are enriched in extracellular matrix remodeling, inflammatory signaling, and fibrogenic pathways associated with hepatic stellate cell activation. Collectively, these findings suggest that GLP-1 RA therapy may improve pediatric MASLD by attenuating fibrogenic and inflammatory transcriptional programs. Although limited by a small cohort size, this integrated clinical-transcriptomic approach supports further investigation of GLP-1 receptor agonists as a therapeutic strategy for pediatric MASLD.

Abstract: Despite advances in engineering, fire safety improvements have plateaued in developed nations, necessitating a reassessment of resource allocation. This study develops a comprehensive fire safety assessment model for the Polish context using the Analytic Hierarchy Process (AHP). A panel of ten experts—comprising fire safety inspectors, State Fire Service officers, and architects—evaluated safety through a two-dimensional framework: the Fire Hazard Index (FHI) and Fire Safety Index (FSI). The results reveal a critical asymmetry: human factors (0.228) and combustible materials dominate the hazard landscape, whereas intelligent AI/IoT systems (0.4133) and passive protection (0.2113) offer the highest potential for safety enhancement. A novel "convergence-divergence" phenomenon was identified: hazard-focused assessments produce convergent priorities across building types (span 0.116), implying universal mitigation needs (e.g., education), while protection-focused assessments yield divergent priorities (span 0.250), justifying targeted investment. Specifically, healthcare facilities (ZL II) require disproportionate protection investment (priority 0.310). The study concludes that sustainable fire safety strategies must combine universal hazard mitigation with targeted technological interventions, offering a data-driven framework for policy optimization in Poland.

Abstract: Ecosystems can undergo abrupt, often irreversible transitions between alternative states —phenomena termed critical transitions or regime shifts— with profound consequences for biodiversity, ecosystem services, and human well-being. Early warning signals (EWS) derived from time series analysis offer the prospect of anticipating such transitions before they occur, potentially enabling preventive management intervention. This review provides a comprehensive synthesis of EWS methods for ecological systems, encompassing theoretical foundations, statistical indicators, empirical applications, and emerging methodological frontiers. We examine the dynamical basis of EWS in critical slowing down theory, wherein systems approaching bifurcation points exhibit characteristic statistical signatures including rising autocorrelation, increasing variance, and spectral reddening. We present a systematic overview of proposed indicators (Table 1), discuss moving-window frameworks for their computation, and critically evaluate preprocessing requirements and sensitivity to analytical choices. Empirical applications across major ecosystem types---including lakes, coral reefs, grasslands, forests, and marine fisheries---reveal both successes and limitations, with EWS performance depending critically on data quality, transition mechanism, and system-specific dynamics (Table 2). We address recent advances including machine learning approaches, non-equilibrium thermodynamic indicators, multivariate extensions, and the important distinction between bifurcation-induced, noise-induced, and rate-induced tipping. We conclude with recommendations for specialists, emphasizing the integration of EWS within broader monitoring frameworks, systematic sensitivity analysis, and the interpretation of indicators as probabilistic assessments of changing resilience rather than deterministic predictions of imminent collapse.

Abstract: We present an information-theoretic framework that models creative reasoning as structured, task-directed motion within hierarchically organized epistemic structures. Creative problem solving is described as a progressive reconfiguration of an initial epistemic state toward a desired target configuration through a sequence of intermediate representations. Each representation is modeled by an empirical structural distribution over extracted features, enabling two complementary quantitative diagnostics: (i) a divergence measure—the Jensen–Shannon (JS) metric—capturing structural departure, novelty, and analogical proximity; and (ii) energy-based plausibility measures expressing conformity to dominant structural regularities and agent-relative constraints. Their interaction induces a geometry in which exploration balances novelty against structural admissibility, and cross-domain transfer is enabled through alignment of compatible representations. We introduce algebraic and probabilistic principles governing the generation, evaluation, and selection of candidate representations, including neighborhood-restricted exploration, history-sensitive evaluation, and non-redundant comparison under progressively refined interpretive conditions. The framework is operationalized at the level of the epistemic structures accessible to an individual reasoning agent, while large language models are interpreted as mechanisms that facilitate access to broader reservoirs of structured knowledge. Although a musical case study (J. S. Bach’s The Art of Fugue) is used for illustration, the proposed framework is domain-general and applies to any setting involving structured representations and lawful transformations. The resulting formalism supports principled approaches to task-oriented creative search, analogical reasoning, and autonomous knowledge exploration, with potential implications for machine-assisted discovery and structurally grounded communication across intelligent systems.

Abstract: Accelerating sustainable food system transitions requires spatially explicit integration of localproduction conditions and nutritional priorities, yet such assessments remain scarce, particularlyfor low- and middle-income countries. We developed an open-source, reproducible nutritionalLife Cycle Assessment model – Local Environmental and Nutritional Scoring (LENS) – andanalyzed sub-national food supply chains across six environmental impact categories in Kenyaand Rwanda. Results reveal strong context dependency: terrestrial animal products showcomparable impacts to most plant-source foods when comprehensively assessed. Enviro-nutritional efficiencies tend to be highest for wild-caught fish and seafood, pulses, fruit andvegetables from low-input systems, and lowest for starchy staples and poultry. Substantialvariation within food groups, between co-products, and across space necessitates interpretingscores at landscape level rather than as independent benchmarks for scaling production.

Abstract: Chronic kidney disease (CKD) is a major global health burden associated with substantially increased risks of morbidity and mortality. Cardiovascular disease remains the leading cause of death across all stages of CKD. Over the past decades, several pharmacologic therapies—including renin–angiotensin system inhibitors, sodium–glucose cotransporter-2 inhibitors, mineralocorticoid receptor antagonists, glucagon-like peptide-1 receptor agonists, and lipid-lowering agents—have demonstrated substantial cardio-nephroprotective benefits and are recommended in international guidelines. However, real-world implementation of these therapies remains incomplete, and emerging evidence highlights important sex-based disparities in prescribing patterns. Although CKD is more prevalent in women worldwide, women with CKD are consistently less likely than men to receive guideline-directed cardioprotective and nephroprotective medications. This treatment gap spans both traditional therapies, such as angiotensin-converting enzyme inhibitors and statins, and newer agents with proven outcome benefits. Women are less likely to initiate treatment, less likely to receive high-intensity or target doses, and less likely to achieve recommended blood pressure and lipid goals. Importantly, the presence of CKD attenuates the usual female survival advantage, and the relative excess cardiovascular risk associated with CKD may be particularly pronounced in women. The under-prescription of cardio-renal therapies in women with CKD reflects a complex interplay of factors. These include older age at presentation, higher reported rates of adverse drug reactions, concerns regarding tolerability and safety in advanced kidney disease, therapeutic inertia, underestimation of cardiovascular risk, and persistent underrepresentation of women in clinical trials. Biological differences in pharmacokinetics and pharmacodynamics, as well as structural and system-level barriers, further contribute to inequities in care. Addressing these disparities requires improved risk recognition, sex-informed prescribing practices, enhanced representation of women in clinical research, and implementation strategies that incorporate sex-disaggregated performance metrics. Reducing treatment gaps is essential to improving cardiovascular and renal outcomes and to achieving equitable, precision-based care for women with CKD.

Abstract: Background: The decline in physical activity during the transition to early adolescence poses a significant threat to lifelong health and well-being, directly impacting the targets of Sustainable Development Goal 3 (SDG 3). To design effective preventive interventions, researchers need developmentally appropriate tools to measure the psychological drivers of physical activity. Objectives: This study aimed to adapt and validate the Attitude To-wards Sport Scale (ATSS), originally developed for high school students, for a middle school population (ages 10–15). Methods: We used a mixed-methods approach, starting with cognitive think-aloud protocols to ensure semantic suitability, followed by a cross-sectional survey of 531 students. Data were analyzed using robust Confirmatory Factor Analysis (CFA). Results: The results confirmed that the original three-factor struc-ture (Interest, Lifestyle, and Participation) perfectly fit the early adolescent sample. The scale demonstrated high composite reliability across all dimensions. Furthermore, the adapted ATSS showed strong criterion-related validity through high correlations with perceived physical literacy and actual physical activity durations. It also successfully dif-ferentiated between licensed athletes and non-licensed students. Conclusions: We con-clude that the adapted ATSS is a highly reliable and developmentally sensitive screening tool for pediatricians, educators, and public health professionals to monitor youth sports engagement and promote sustainable health-lifestyle behaviors.

Abstract: Rheumatoid arthritis (RA) is a chronic inflammatory disease of autoimmune background and unknown etiology. The importance of genetic factors in the RA development is well established. Environmental factors have also been extensively researched in relation to risk of RA and managing its symptoms. Smoking, physical activity, diet and gut microbiota are considered to be the most essential modifiable factors in RA. Among dietary interventions the most researched is Mediterranean diet, monounsaturated fatty acids, fish consumption and fish oil (EPA, eicosapentaenoic acid and DHA, docosahexaenoic acid). Others concerned gluten-free and vegan or vegetarian diet, salt intake, supplementation with vitamin D, antioxidants, prebiotics and probiotics. Diet modifications can alter the gut environment and the association between RA development or severity and composition of gut bacteria has already been shown. This review focuses on effectiveness and usefulness of various dietary approaches and supplements in RA prevention and management, including influence on disease activity and inflammatory status. Composition of gut microbiota and its changes in response to dietary factors are also considered. There is a great need for further research into mutual dependencies of diet, microbiome and RA activity. The current state of knowledge provides promising evidence for future nutrition and microbial therapies.

Abstract: Background. The preoperative differential diagnosis of myometrial lesions remains a significant challenge when using conventional imaging techniques, such as ultrasound (US) and magnetic resonance imaging (MRI). Radiomics and machine learning, which leverage quantitative features beyond human visual perception, are increasingly recognized as promising tools for improving differential diagnosis in gynecology. Methods. This retrospective study included patients who underwent surgery for uterine masses and had preoperative MR. A machine learning model was developed to analyze radiomic features extracted from T2-weighted and diffusion-weighted MR images. Results. 44 subjects were included: 19 (43.2%) classified as "sarcoma" and 25 (56.8%) as "fibroid" based on histology after surgery. This dataset was used for training and cross-validation of different models. Three models, comprising ensembles of machine learning classifiers (random forests, support vector machines, and k-nearest neighbors), were developed for binary classification using histological diagnosis as reference standard. The best-performing model achieved the following results: AUC 90%, accuracy 82%, sensitivity 95%, specificity 72%, PPV 72%, and NPV 95%. Conclusions. Our model demonstrated high sensitivity and moderate accuracy, suggesting its potential as a valuable tool for assisting clinicians in the preliminary assessment of myometrial lesions and guiding decision-making toward conservative management in cases of non-suspicious masses.

Abstract: Urban Embodied Agents (UrbanEAs) are emerging to actively interact with complex, large-scale city environments and generate vast, heterogeneous data streams, moving beyond the single-vehicle of existing autonomous driving. However, urban environments present distinct challenges, including environmental variability, limited observability, and interaction complexity. These challenges hinder the effectiveness of existing embodied agents, which have focused on controlled indoor environments, and expose the inherent limitations of relying on single-domain data. Therefore, establishing a comprehensive data lifecycle to fuse multidomain data from terrain, aerial, and space is a strategy for developing actionable embodied capabilities from raw urban streams. Distinct from existing surveys that follow a model-centric paradigm for urban computing or autonomous driving, we systematically propose and review a comprehensive Data Lifecycle from a multidomain data perspective, which is critical for the UrbanEA. First, we propose a unified framework containing four key stages of this lifecycle: Data Perception, Data Management, Data Modeling, and Task Application. Next, we establish a taxonomy for each stage of the lifecycle. Specifically, we detail the evolution from static data storage to active agent memory, and analyze integration strategies designed to bridge multidomain gaps. We demonstrate how UrbanEAs empower downstream tasks, including Urban Scene Question-Answering (SQA), Vision-Language Navigation (VLN), and Human-Agent Collaboration (HAC). Finally, we outline the social impact of the data lifecycle of UrbanEA and open research problems with the future directions. Our survey provides a roadmap for designing the robust, high-performance data frameworks essential for these UrbanEAs.