Search Results (29,606)

Understanding how herbivores adjust their foraging strategies to cope with seasonal resource fluctuations has been central to the nutritional ecology. Optimal Foraging Theory (OFT) predicts that generalists should broaden their dietary niche when high-quality resources are scarce, but empirical evidence in extreme environments remains poorly understood. We used trnL-P6 metabarcoding of fecal samples (n = 10/season) and a local reference library of 120 plant species to quantify diet composition and niche metrics of free-ranging yaks (Bos grunniens) on the Qinghai–Tibetan Plateau in June (summer) and October (autumn) 2024. Yaks shifted from a diverse, forb-dominated diet (e.g., Polygonaceae, Rosaceae) in summer to a specialized diet dominated by grasses in autumn. Although dietary richness and total niche width (TNW) decreased in autumn, phylogenetic diversity remained stable, indicating a strategic shift to distinct evolutionary lineages to ensure functional redundancy. Furthermore, food network analyses demonstrated a transformation from a flexible, modular foraging pattern in summer to a highly integrated, synchronized network in autumn. These findings suggest that under the distinct quality–quantity trade-off of high-altitude ecosystems, yaks adopt an energy-maximization strategy by minimizing search costs, aligning with the opportunity cost constraints of OFT, rather than randomly expanding their niche. This insight into selective foraging dynamics is critical for developing sustainable grazing practices that accommodate the natural adaptive behaviors of alpine herbivores. Full article

In response to the environmental and health concerns associated with high-sodium brine disposal and the sodium content in table olives, this study proposes a novel, sustainable preservation method that completely replaces traditional brine with chitosan solutions. Three food-grade chitosan solutions were formulated using acetic acid, vinegar, and vinegar neutralized with baking soda as alternative liquid media for preserving Kalamata olives. Over a five-month storage period with a one-year endpoint, these solutions were evaluated against a conventional 8% NaCl brine control. The chitosan-based systems demonstrated effective microbial control, maintaining significantly lower total viable counts for most of the storage period, while yeast and mold populations were comparable to or slightly higher than the control over extended storage. Notably, they reduced the medium’s salinity by 75–85%, directly addressing the issue of high sodium content. The chitosan solutions also provided superior pH stability and color maintenance in the olives. A key finding was the distinct nature of the interaction between the olives and the chitosan medium compared to brine: while antioxidant activity within the olive flesh declined, the chitosan solutions themselves exhibited high and stable intrinsic antioxidant capacity (>78%), acting as an active antioxidant reservoir—a dynamic not observed with traditional brine. This research successfully validates chitosan solution as a viable, low-sodium, brine-free preservation medium, offering a novel strategy for sustainable olive processing that valorizes seafood waste and aligns with circular economy principles. Full article

Lymphatic filariasis (LF) is a mosquito-borne neglected tropical disease targeted by the World Health Organization (WHO) for global elimination as a public health problem. Sixteen Pacific Island countries and territories were historically endemic, and eight have now met the WHO criteria for elimination as a public health problem. Elimination as a public health problem does not imply zero transmission. Rather, it denotes that LF prevalence has been reduced below a defined threshold at which community transmission can be sustained. Following validation of elimination, the WHO recommends post-validation surveillance (PVS) to detect potential re-emergence of LF as a public health problem. However, implementing PVS is challenging in Small Island Developing States with dispersed populations, limited workforce capacity, resource constraints, and competing health priorities. The ‘Voices and Visions: Building Partnerships for Integrated Serosurveillance of LF and Other Infectious Diseases in the Pacific Islands’ meeting was held in Brisbane, Australia, from 8–10 July 2025. Fifty-one delegates, including Pacific LF programme managers, WHO representatives, global health partners, and academic researchers, reviewed regional PVS progress, discussed the newly released WHO guidelines for the implementation, monitoring, and evaluation of PVS, planned for PVS implementation, and explored novel multiplex bead assay (MBA) serological analysis methods to strengthen regional coordination for its development as a public health tool. Five broad themes emerged. First, the new WHO Monitoring and Epidemiological Assessment of Mass Drug Administration in the Global Programme to Eliminate Lymphatic Filariasis: A Manual for National Elimination Programmes, 2nd edn needs to be operationalised to meet decision-making needs across diverse Pacific settings. Second, integrating LF-PVS with existing surveys and health service activities could improve efficiency and long-term sustainability. Third, regional coordination and alignment of funding cycles will require high-level collaboration. Fourth, community engagement is essential to strengthen demand for PVS. Finally, while at an early stage and with further evidence needed, MBA laboratory methods hold promise for cost-effective, feasible integrated multi-pathogen serosurveillance. Full article

In the post-pandemic era, student mental health challenges have emerged as a critical issue in higher education. However, conventional assessment approaches often treat at-risk populations as a monolithic entity, thereby limiting intervention effectiveness. This study proposes a novel computational framework that integrates explainable artificial intelligence (XAI) with unsupervised learning to decode the latent heterogeneity of psychological risk mechanisms. We developed a “predict-explain-discover” pipeline leveraging TreeSHAP and Gaussian Mixture Models to identify distinct risk subtypes based on a 2556-dimensional feature space encompassing lexical, linguistic, and affective indicators. Our approach identified three theoretically-grounded subtypes: academically-driven (28.46%), socio-emotional (43.85%), and internal regulatory (27.69%) risks. Sensitivity analysis using top-20 core features further validated the structural stability of these mechanisms, proving that the subtypes are anchored in the model’s primary decision drivers rather than high-dimensional noise. The framework demonstrates how black-box classifiers can be transformed into diagnostic tools, bridging the gap between predictive accuracy and mechanistic understanding. Our findings align with the Research Domain Criteria (RDoC) and establish a foundation for precision interventions targeting specific risk drivers. This work advances computational mental health research through methodological innovations in mechanism-based subtyping and practical strategies for personalized student support. Full article

The rapid expansion of tourism in Bali and Lombok has precipitated a state of overtourism, critically challenging their ecological and socio-cultural carrying capacities. This study, conducted between 2023 and 2024, employs a qualitative case study approach to investigate the manifestations of overtourism and the efficacy of prevailing mitigation strategies. Data were collected through 32 in-depth interviews, four focus group discussions, and extensive field observations across key destinations in both islands. The findings reveal that overtourism is not merely a function of high visitor numbers but a symptom of systemic governance failure. Key manifestations include acute environmental degradation, the commodification of cultural heritage, and significant economic leakage that marginalizes local communities. These issues are exacerbated by fragmented policy, weak regulatory enforcement, and the exclusion of local voices from tourism planning. The study concludes that technical solutions such as visitor quotas are insufficient without a fundamental governance paradigm shift. Effective mitigation requires an integrated approach centered on strict carrying capacity enforcement, genuine community empowerment through Community-Based Tourism (CBT), and the strategic use of digital tools for visitor dispersion. This research provides an empirically grounded framework that underscores the imperative of a fundamental governance paradigm shift, aligning tourism development in island destinations with the principles of sustainability and equity. Full article

We investigate how symmetries present in datasets affect the structure of the latent space learned by Variational Autoencoders (VAEs). Understanding symmetries in data is essential because symmetries determine the true degrees of freedom, constrain generalization, and provide physically interpretable coordinates. We therefore study whether a standard, non-equivariant VAE can reveal symmetry-induced dimensional reduction directly from data, without imposing the symmetry in the architecture. By training VAEs on data originating from simple mechanical systems and particle collisions, we analyze the organization of the latent space through a relevance measure that identifies the most meaningful latent directions. We show that when symmetries or approximate symmetries are present, the VAE self-organizes its latent space, effectively compressing the data along a reduced number of latent variables. This behavior captures the intrinsic dimensionality determined by the symmetry constraints and reveals hidden relations among the features. Furthermore, we provide a theoretical analysis of a simple toy model, demonstrating how, under idealized conditions, the latent space aligns with the symmetry directions of the data manifold. We illustrate these findings with examples ranging from two-dimensional datasets with $O\left(2\right)$ symmetry to realistic datasets from electron–positron and proton–proton collisions. Our results highlight the potential of unsupervised generative models to expose underlying structures in data and offer a novel approach to symmetry discovery without explicit supervision. Full article

Stance detection has emerged as an essential tool in natural language processing for understanding how individuals express agreement, disagreement, or neutrality toward specific targets in social and online discourse. It plays a crucial role in bilingual and multilingual environments, including English-Arabic social media ecosystems, where differences in language structure, discourse style, and data availability pose significant challenges for reliable stance modelling. Existing approaches often struggle with target awareness, cross-lingual generalization, robustness to noisy user-generated text, and the interpretability of model decisions. This study aims to build a reliable, explainable target-aware bilingual stance-detection framework that generalizes across heterogeneous stance formats and languages without retraining on a dataset specific to the target language. Thus, a unified dual-encoder architecture based on mDeBERTa-v3 is proposed. Cross-language contrastive learning offers an auxiliary training objective to align English and Arabic stance representations in a common semantic space. Robustness-oriented regularization is used to mitigate the effects of informal language, vocabulary variation, and adversarial noise. To promote transparency and trustworthiness, the framework incorporates token-level rationale extraction, enables fine-grained interpretability, and supports analysis of hallucination. The proposed model is tested on a combined bilingual test set and two structurally distinct zero-shot benchmarks: MT-CSD and AraStance. Experimental results show consistent performance, with accuracies of 85.0% and 86.8% and F1-scores of 84.7% and 86.8% on the zero-shot benchmarks, confirming stable performance and realistic generalization. Ultimately, these findings reveal that effective bilingual stance detection can be achieved via explicit target conditioning, cross-lingual alignment, and explainability-driven design. Full article

Speech-driven face generation aims to synthesize a face image that matches a speaker’s identity from speech alone. However, existing methods typically trade identity fidelity for visual quality and rely on large end-to-end generators that are difficult to train and tune. We propose Vox2Face, a speech-driven face generation framework centered on an explicit identity space rather than direct speech-to-image mapping. A pretrained speaker encoder first extracts speech embeddings, which are distilled and metric-aligned to the ArcFace hyperspherical identity space, transforming cross-modal regression into a geometrically interpretable speech-to-identity alignment problem. On this unified identity representation, we reused an identity-conditioned diffusion model as the generative backbone and synthesized diverse, high-resolution faces in the Stable Diffusion latent space. To better exploit this prior, we introduce a discriminator-free diffusion self-consistency loss that treats denoising residuals as an implicit critique of speech-predicted identity embeddings and updates only the speech-to-identity mapping and lightweight LoRA adapters, encouraging speech-derived identities to lie on the high-probability identity manifold of the diffusion model. Experiments on the HQ-VoxCeleb dataset show that Vox2Face improves the ArcFace cosine similarity from 0.295 to 0.322, boosts R@10 retrieval accuracy from 29.8% to 32.1%, and raises the VGGFace Score from 18.82 to 23.21 over a strong diffusion baseline. These results indicate that aligning speech to a unified identity space and reusing a strong identity-conditioned diffusion prior is an effective method to jointly improve identity fidelity and visual quality. Full article

Thermal interface materials are critical components for ensuring efficient heat dissipation in thermal management systems. The current research focus is to fabricate thermal interface materials (TIMs) that demonstrate high thermal conductivity while at low filler loadings. In this study, an aligned, thermally conductive skeleton was fabricated via the freeze casting method, utilizing carbon nanofibers (CNFs) and nickel (Ni) particles. This skeleton was subsequently infiltrated with silicone rubber (SR) to obtain the polymer composite. Within the aligned skeleton, CNFs and Ni particles are densely packed, with the Ni particles acting as conductive bridges between adjacent CNFs. This bridging effect facilitates a substantial enhancement in the overall thermal conductivity with only a minimal addition of Ni. By combining the skeleton’s microstructure with thermal performance, the effects of key parameters on thermal conductivity were systematically investigated. A maximum thermal conductivity improvement of 64.8% was achieved by hybridizing CNFs with a small amount of Ni (1.09 vol%) compared to the CNF-only counterpart. Furthermore, at a low total loading (8.02 vol% CNFs and 1.09 vol% Ni), the composite achieved a thermal conductivity of 3.30 W/(m·K). This value was 47.2% higher than that of a CNF-only TIM and 36.2% higher than that of a composite prepared by common freezing under the same filler composition. Additionally, the incorporation of Ni enhanced the composite’s thermal stability. Moreover, the composite exhibited a favorable combination of enhanced mechanical strength and excellent elasticity. Full article

Background: Large language models (LLMs) are increasingly proposed as clinical decision support tools. However, their reliability in the emergency department (ED) triage remains insufficiently validated. This study aimed to evaluate the performance and limitations of multiple LLMs in triage using a large retrospective dataset. Methods: We conducted a retrospective analysis of 39,375 anonymized patient cases from the ED of AHEPA University General Hospital, Thessaloniki, Greece (June 2024–July 2025), extracted from the hospital’s electronic medical record system. All cases were triaged in real time according to the Emergency Severity Index (ESI) by 25 emergency physicians. In cases of uncertainty, a senior emergency physician was consulted. Seven LLMs (ChatGPT-5 Thinking, ChatGPT-5 Instant, Gemini 2.5, Qwen 3, Grok 4.0, Deep Seek v3.1, and Claude Sonnet 4) were evaluated against the physician-assigned ESI level (reference standard). Outcomes included triage score agreement (quadratic weighted kappa, κw), clinic referral accuracy and admission prediction. Subgroup analyses were performed by referral clinic and admission outcome. The study was conducted in accordance with TRIPOD-AI reporting guidelines. Results: Model performance varied substantially. DeepSeek and Claude Sonnet 4 achieved the highest agreement with physician-assigned ESI (κw ≈ 0.467; raw accuracy: 61.7%). In contrast, GPT-5 Instant performed poorly across all evaluation metrics (κw = 0.176; 95% CI: 0.167–0.186). Claude Sonnet 4 demonstrated the best performance in clinic referral (67.1%; κ = 0.619) and admission prediction (κw ≈ 0.46). Subgroup analyses indicated higher performance in pediatric cases and organ-specific complaints, such as ophthalmology (up to 81% accuracy). LLMs also showed tendencies toward over- or under-triage. Conclusions: Current LLMs demonstrate promising but inconsistent capability in triage. While selected models achieved moderate alignment with physician ESI decisions, none achieved strong agreement (κ > 0.80). LLMs are most suitable as supervised decision support tools, particularly in anatomically well-defined clinical scenarios, rather than as autonomous systems. Full article

The escalating demand for global fruit logistics underscores the urgency of packaging innovations to reconcile preservation efficiency with environmental sustainability, particularly addressing microplastic pollution from conventional plastics and safety hazards posed by synthetic pH-sensitive pigments. Natural pH-sensitive intelligent edible gel-based packaging, which integrates non-toxic indicators into biopolymer gel matrices, offers a viable solution by visually tracking freshness through colorimetric responses to pH fluctuations during storage and transportation. This review systematically synthesizes recent progress in material design, including the development of edible films and coatings, and evaluates the functional mechanisms of natural pH indicators within these systems. Applications across diverse fruit categories demonstrate their efficacy in delaying ripening, inhibiting microbial growth, and signaling quality degradation via dynamic color shifts. Despite enabling real-time, visual freshness monitoring, challenges in mechanical robustness, water resistance, and scalable manufacturing remain. Future advancements should prioritize the integration of multifunctional systems, such as gas conditioning technologies and bioactive components, to enhance practical performance and align with sustainable food preservation objectives, ultimately reducing food waste and elevating consumer safety standards. Full article

Background: In the Americas, the number of people living with diabetes is expected to rise from 92 million in 2024 to 120 million by 2050. Indigenous populations may experience distinct biological, environmental, and sociocultural risk factors; however, they are often treated as a homogeneous group in epidemiological research, and consolidated evidence on diabetes prevalence across diverse Indigenous populations remains limited. This scoping review examines the prevalence of diabetes among Indigenous populations in the Americas. Methods: Following PRISMA-ScR guidelines, we conducted a systematic scoping review of population-based studies reporting the prevalence of diabetes among Indigenous adult populations in the Americas. Searches were performed in PubMed and Scopus. Collected data included study location, Indigenous group, population characteristics, diagnostic criteria, and test used and reported prevalence estimates. Results: Sixty documents encompassing 73 studies met the inclusion criteria, representing 45,503 individuals from 16 countries between 1975 and 2025. The total number of ethnic groups represented was 111, and 12 studies did not identify a specific ethnic group. Fasting blood glucose (FBG) was the most frequently used diagnostic method, followed by the oral glucose tolerance test (OGTT). Estimates of the prevalence of diabetes varied widely across populations, regions, and time periods. Five studies—from Brazil, Chile, Colombia, Mexico, and Paraguay—did not identify any cases of diabetes. Among studies reporting cases, prevalence ranged from 1 to 70% in North America, 5 to 14% in Central America, and 1 to 29% in South America. Conclusions: The prevalence of diabetes among Indigenous populations varied widely across the region, with substantially higher estimates reported in North America than in Central and South America. The decline in published studies in recent years suggests reduced research attention to this topic. The marked heterogeneity identified in this review underscores the need for standardized measurement approaches to support population-specific strategies aligned with personalized care and precision public health. Full article

Studying the mechanisms by which Gram-negative heterotrophic bacteria transition from active metabolism to dormancy is an important task, as it is directly related to the problem of bacterial antibiotic resistance and the spread of nosocomial infections. Using electron microscopy, microbiology, and molecular modeling, we investigated the dose-dependent mechanisms of action of 4-hexylresorcinol (4HR), a chemical analog of the anabiosis autoinducer, on the cell membranes of Gram-negative bacteria (using Escherichia coli as an example), leading to the formation of stressed, dormant, and mummified cells. It was shown that 4HR penetrates membranes equally easily both as single molecules and as micelles, distributing itself across the membrane so that the hydrocarbon radicals are aligned parallel to the lipid tails. When micelles penetrate the membrane, uneven distribution of 4HR within and between leaflets occurs, as well as lipid redistribution within the membrane, leading to the appearance of a third peak on the phospholipid electron density profile and a third black band in the membrane region in TEM images of such cells. At 4HR concentrations in solution of 200 µM, its micelles cover the cell membranes in a thick layer, penetrate into the membrane, and completely saturate it. Even higher concentrations create agglomerates or actually micellar arrays within the cell membranes, leading to cell death through mummification. Full article

Cross-domain sentiment analysis remains challenging due to distributional shifts and heterogeneous sentiment expressions across platforms. Existing domain adaptation approaches primarily focus on enforcing domain-invariant representations. However, such symmetry-preserving strategies often overlook directional and expression-level asymmetries. These asymmetries naturally arise in real-world sentiment data, particularly for context-inferred sentiment expressions. In this work, we propose a novel symmetry- and asymmetry-aware domain adaptation framework for cross-domain sentiment classification. The framework models symmetry through explicit multi-source distribution alignment, which captures transferable sentiment knowledge across domains. Additionally, aspect-level structural supervision organizes representations according to shared linguistic aspects. To address asymmetry, a directional divergence regularization is introduced. This component models expression-level and directional discrepancies between source and target domains. Importantly, the framework operates without requiring target-domain annotations. Experiments are conducted under a multi-source unsupervised domain adaptation setting using sentence-level hotel review datasets collected from multiple online platforms. Empirical results demonstrate strong performance for the proposed framework. It achieves an average Accuracy of 82.0% and Macro-F1 of 80.6%. The framework consistently and statistically significantly outperforms source-only, multi-source, and transformer-based adversarial adaptation baselines across all evaluated target domains (p < 0.05). Additional analyses confirm improved robustness to implicit sentiment expressions and platform-induced asymmetries. These findings highlight the importance of jointly modeling symmetry and asymmetry for robust cross-domain sentiment adaptation and provide a unified and deployable solution for sentiment analysis under realistic platform shifts. Full article

Metabolic dysfunction-associated steatotic liver disease (MASLD) is an emerging global health challenge with limited effective therapeutic options. The gut microbiota, at the interface of host metabolism and immunity, acts as a critical disease modifier via the gut–liver axis. This review goes beyond cataloging its associations and synthesizes how intrinsic and extrinsic factors sculpt a permissive microbial ecosystem. These factors likely converge to establish a state of “metabolic dysbiosis”, fueling MASLD progression through three core mechanisms: compromised intestinal barrier integrity with immune activation, dysregulation of key microbial metabolite axes, and direct hepatic insult from gut-derived products. Next, we evaluate the translational landscape through a mechanism-informed precision framework, with an emphasis on how microbiome-based interventions could be aligned with non-invasive biomarkers increasingly used for MASLD risk stratification and treatment monitoring. By integrating evidence across scales, this review aims to frame a roadmap from microbiome correlations to causality-driven, personalized therapeutic strategies for MASLD. Full article