Search Results (288)

Large Language Models (LLMs) have become foundational to modern Artificial Intelligence (AI), enabling advanced reasoning, multimodal understanding, and scalable human-AI interaction across diverse domains. This survey provides a comprehensive review of major proprietary and open-source LLM families, including GPT, LLaMA 2, Gemini, Claude, DeepSeek, Falcon, and Qwen. It systematically examines architectural advancements such as transformer refinements, mixture-of-experts paradigms, attention optimization, long-context modeling, and multimodal integration. The paper further analyzes alignment and safety mechanisms, encompassing instruction tuning, reinforcement learning from human feedback, and constitutional frameworks, and discusses their implications for controllability, reliability, and responsible deployment. Comparative analysis of training strategies, data curation practices, efficiency optimizations, and application settings highlights key trade-offs among scalability, performance, interpretability, and ethical considerations. Beyond synthesis, the survey introduces a structured taxonomy and a feature-driven comparative study of over 50 reconstructed LLM architectures, complemented by an interactive visualization interface and an open-source implementation to support transparency and reproducibility. Finally, it outlines open challenges and future research directions related to transparency, computational cost, data governance, and societal impact, offering a unified reference for researchers and practitioners developing large-scale AI systems. Full article

As the electrification of transportation systems accelerates, incentivizing electric vehicle (EV) participation in vehicle-to-grid (V2G) operations is becoming increasingly crucial. This paper introduces a dynamic EV charging and discharging strategy that incorporates integrated energy station (IES) congestion and electricity purchase and sale scenarios. The proposed strategy seeks to facilitate orderly EV charging and discharging within a real-time simulation framework that integrates the transportation network (TN), IES, and the external grid (EG). First, we develop a real-time collaborative simulation framework that combines microscopic traffic flow (MTL) and IES–grid energy interaction models to account for mutual feedback among these components. Second, we propose an EV IES selection strategy aimed at maximizing discharge revenue, which takes into account various factors, including driving distance, time costs, battery degradation, discharge benefits, and government subsidies. Finally, we design a dynamic discharge pricing model based on real-time vehicle arrival patterns at the IES and the status of electricity purchases and sales. Simulation results show that the EV IES selection strategy, optimized for discharge revenue, reduces average user waiting time by 5.36%, decreases network time loss by 3.86%, and increases EV discharge revenue by 6.79%. Furthermore, the introduction of dynamic pricing leads to additional reductions in waiting time and network time loss by 3.46% and 4.80%, respectively. The proposed mechanism and pricing strategy effectively mitigate traffic congestion, enhance user discharge revenue, and provide flexible scheduling options for IES operations. Full article

This study examines how firms in low- and medium-technology (LMT) manufacturing industries improve product stability during technological upgrading. While existing innovation research has largely focused on high-technology sectors and radical innovation, relatively little attention has been paid to stability-oriented innovation in traditional manufacturing industries. In machinery manufacturing, however, product reliability and operational stability are often more critical to customers than frequent technological novelty. To address this gap, the study investigates the mechanisms through which firms enhance product stability in the wood-based panel machinery industry. The research adopts a longitudinal multiple-case study approach based on two leading Chinese machinery manufacturers located in the Linyi industrial cluster. Empirical data were collected through in-depth interviews, field observations, and secondary materials. The findings indicate that product stability improvements emerge from the interaction of industrial agglomeration, incremental innovation, and technological integration. Industrial clusters facilitate knowledge exchange among firms, suppliers, and downstream users. Continuous incremental improvements based on operational feedback enable firms to refine product designs, while technological integration allows them to combine internal capabilities with external technological components. By highlighting the importance of stability-oriented innovation, this study contributes to the literature on innovation in LMT industries and provides insights into the technological learning processes of machinery manufacturers. The findings also offer practical implications for managers and policymakers seeking to promote technological upgrading in traditional manufacturing sectors. Full article

This study focuses on the commonalities and differences in the public housing systems of three East Asian countries, using Japan’s UR Rental Housing as a case study. It employs a composite methodology that integrates architectural typology and cross-cultural comparison, constructing theoretical linkages within a three-dimensional framework of “social institutions–cultural context–spatial structure”. The research emphasizes three key dimensions: (1) The evolution of policy frameworks and their underlying socio-cultural drivers; (2) The spatial layout logic and functional concepts embedded in residential unit planning; (3) The transformation and inheritance of traditional residential values in contemporary housing design. The study strictly adheres to a progressive logic of “sample construction–type decoding–paradigm extraction–cross-domain comparison–theoretical feedback”. It begins by analyzing the core issues in the supply structure and spatial adaptability of affordable housing in China and South Korea. Next, it systematically examines the policy evolution and spatial design paradigms of Japan’s UR Rental Housing. Subsequently, it constructs a comparative analytical matrix for public housing in China, Japan, and South Korea, identifying transferable common experiences and pathways requiring localized adaptation. Finally, it proposes targeted recommendations across three dimensions, namely policy framework, spatial design, and community building: (1) At the policy level, a full lifecycle governance framework is advocated; (2) In spatial design, the principles of “compactness and efficiency” are emphasized, alongside enhanced flexibility and cultural relevance; (3) In community building, efforts are directed toward activating interpersonal connections and strengthening the social functional attributes of housing. This study emphasizes transnational comparability and knowledge transferability, aiming to provide practical insights for China’s affordable housing reforms and South Korea’s public housing modernization. It seeks to promote cross-national learning and collaborative innovation in the regional housing sector, offering both theoretical reference and practical pathways to realize the shared vision of “restoring housing to a human scale”. Full article

Subway public spaces have been identified as a vital medium for showcasing urban culture. The design quality of these spaces has been shown to have a profound influence on passengers’ spatial perception and cultural experience. However, amid rapid urbanization, subway stations commonly face issues such as homogeneous spatial interfaces and unclear cultural themes, resulting in diminished station identifiability. This study integrates post-use evaluation with Importance–Performance Analysis (IPA) to establish an assessment and optimization pathway aimed at systematically identifying and prioritizing key design elements for enhancing cultural identifiability. Taking Tianjin Gulou Station as a case study, user feedback collected through questionnaires identified 12 indicators influencing identifiability satisfaction. The reliability and validity of the questionnaire were confirmed through validity analysis and paired-sample t-tests, while IPA was employed to clarify improvement priorities. The results indicate that the overall perceived importance of cultural identifiability at Gulou Station significantly exceeds satisfaction levels. Landmark installations, art walls, and vertical transportation fall within the “high importance-low satisfaction” quadrant, which is identified as a primary area of focus for enhancement. Basic interface elements such as flooring and ceilings require enhancement, while transfer entrances and station name walls constitute advantageous designs warranting preservation. Based on the findings of the present study, three targeted design strategies are proposed: enhancing spatial perception, constructing cultural continuity, and integrating multidimensional experiences. These approaches seek to address the “spatial-cultural” perception gap, providing actionable pathways for the distinctive renewal of subway spaces. Full article

Italian hand gestures constitute a distinctive and widely recognized form of nonverbal communication, deeply embedded in everyday interaction and cultural identity. Despite their prominence, these gestures are rarely formalized or systematically taught, posing challenges for foreign speakers and visitors seeking to interpret their meaning and pragmatic use. Moreover, their ephemeral and embodied nature complicates traditional preservation and transmission approaches, positioning them within the broader domain of intangible cultural heritage. This paper introduces a machine learning–based framework for recognizing iconic Italian hand gestures, designed to support cultural learning and engagement among foreign speakers and visitors. The approach combines RGB–D sensing with depth-enhanced geometric feature extraction, employing interpretable classification models trained on a purpose-built dataset. The recognition system is integrated into a non-immersive virtual reality application simulating an interactive digital totem conceived for public arrival spaces, providing tutorial content, real-time gesture recognition, and immediate feedback within a playful and accessible learning environment. Three supervised machine learning pipelines were evaluated, and Random Forest achieved the best overall performance. Its integration with an Isolation Forest module was further considered for deployment, achieving a macro-averaged accuracy and F1-score of 0.82 under a 5-fold cross-validation protocol. An experimental user study was conducted with 25 subjects to evaluate the proposed interactive system in terms of usability, user engagement, and learning effectiveness, obtaining favorable results and demonstrating its potential as a practical tool for cultural education and intercultural communication. Full article

Australian regional communities are actively seeking development pathways that generate local economic value while maintaining environmental and cultural integrity. In this context, GeoRegions have emerged in Australia as a community-led approach for recognising and interpreting geoheritage and associated abiotic–biotic–cultural (ABC) values through geotourism and geoeducation. The GeoRegion concept remains intentionally operationally flexible, but for regional communities encountering a myriad of barriers to sustainable geotourism implementation, any uncertainty for proponents about what constitutes an implementable GeoRegion and what resources and governance arrangements are required for credible and sustained delivery requires resolution. This study developed a stakeholder-informed conceptual model to clarify the practical ‘building blocks’ of GeoRegion establishment and the conditions under which GeoRegions can contribute to sustainability-oriented regional development. Using a design thinking framing and semi-structured interviews with thirteen expert participants, we used semantic discourse analysis to identify the factors perceived as essential to GeoRegion viability and legitimacy. We found that participants expected GeoRegions to be geologically centred, but their perceived value and long-term durability depend on (i) genuine community support and locally legitimate narratives (including Indigenous knowledge where appropriate), (ii) capable champions or coordinating groups, (iii) sustained resourcing for interpretation and visitor readiness, and (iv) a facilitative and not prescriptive role for government. Participants emphasised that GeoRegions should never be constrained by land tenure but cautioned that competing land uses, access logistics and uneven capacity across regions were highly influential in the delineation of feasible boundaries and management intensity. Our GeoRegion model differentiates core inputs (community mandate, knowledge co-production, geoheritage significance, human capacity and funding) from expected outputs (interpretive materials, geoeducation, geotourism, economic development, conservation outcomes and strengthened place identity), and we identify feedback that can either reinforce or erode sustainability outcomes over time. We argue that GeoRegions can provide a low-risk, scalable mechanism for geoconservation-informed regional development, particularly where formal protected-area tools or geopark ambitions are politically or economically constrained, provided that supporting governance and resourcing are treated as essential design requirements. Full article

This study explores the design and implementation of active learning strategies in a Digital Logic Design (DLD) course offered at Sultan Qaboos University. By shifting away from traditional lecture formats and incorporating tools such as interactive quizzes, collaborative whiteboard sessions, and real-time feedback mechanisms, the course aimed to increase student engagement and deepen conceptual understanding. A mixed-methods approach was used, including classroom observation, visual documentation, and post-course satisfaction surveys (n = 49). Inferential statistical analyses and effect sizes were calculated on key survey items. Findings indicate that these strategies significantly enhanced both cognitive and affective aspects of learning. This paper offers insights for educators in STEM disciplines seeking to cultivate active learning environments that align with contemporary pedagogical frameworks. Full article

The increasing adoption of Large Language Models (LLMs) in higher education has created a need for high-quality, publicly available benchmarks for automated assessment. Existing datasets often rely on synthetic responses or lack detailed human feedback. This paper presents a multilingual dataset of 3041 authentic student answers to 50 open-ended Computer Science questions, collected from real university assessments during the 2025–2026 academic year. The dataset includes the original student responses (Spanish) and their parallel translations (English), instructor (or teacher) defined ideal answers, blind human grading with qualitative feedback, and structured evaluations from three state-of-the-art LLMs (DeepSeek-chat-V3.2, Qwen-flash-2025-07-28, Gemini-2.5-flash-lite-001) using a unified JSON schema. This resource enables reproducible research in automated grading, feedback generation, and cross-lingual educational NLP. Full article

This paper presents a model-free Best Efficiency Point (BEP) tracking method for centrifugal pumps without head measurement or manufacturer-provided characteristic curves. The proposed approach combines a discrete finite-difference extremum-seeking control (ESC) scheme with an efficiency approximation proxy derived from measurable variables—namely, flow rate and electrical power. Under constant head conditions, the proxy function is analytically shown to be proportional to the true pump efficiency, enabling real-time BEP localization using only motor feedback signals. The ESC algorithm employs a sign-based gradient rule with adaptive step-size reduction to achieve rapid and stable convergence without mathematical models. A Python-based simulation using a Schneider SUB 15-0.5cv pump demonstrates that the method can track the BEP with negligible steady-state error (less than 0.1% efficiency deviation). The proposed framework offers a cost-effective solution for efficient optimization for mobile pumping applications in large water resources where installing head sensors is impractical. Full article

The food processing industry is seeking new technologies to enhance product safety, extend shelf life, and optimise food quality in response to growing consumer demand for high-quality products. Since the discovery of plasma technology, its potential applications in food processing have been numerous. For most of these applications, plasma characterisation is key to successfully scaling up from laboratory to industrial settings. A highly valuable tool for plasma characterisation is optical emission spectroscopy (OES), which serves as a non-invasive diagnostic method to monitor reactive species, specifically excited atoms and molecules (reactive oxygen and nitrogen species—RONS) that are critical for food treatment. The main role of OES in food control is to enable species identification and real-time process monitoring, providing feedback on electron temperature and density to prevent thermal damage to sensitive food products. It also facilitates optimisation by adjusting voltage and gas flow rates to maximise the production of antimicrobial species. These results ensure that processes are reliable and repeatable, supporting the transition from laboratory-scale to industrial applications. The paper provides an overview of the use of optical emission spectroscopy in various applications of plasma technology in food processing, including the determination of the elemental composition of raw materials and final products, detection of contaminants, quality control, determination of characteristic plasma parameters, and ensuring compliance with food safety regulations. Full article

Authentic learning is widely recognised as critical for graduate employability, yet embedding it across curricula presents challenges, including resourcing, time for redevelopment, and managing large-scale cultural change. Scalable approaches require pilot studies to explore how authentic learning principles can be implemented and scaffolded effectively. This paper presents a case study of an Australian university’s co-created ‘Authentic Learning Blueprint’, designed to embed industry/community engaged learning throughout the learner’s experience. Phase one involved a team of educators, drawing on insights from learners, and employer feedback to co-develop the Blueprint. The Blueprint allows learners to progress through scaffolded stages, Novice, Associate, and Emerging Practitioner, gaining discipline-specific and industry-ready skills through real-world project briefs and work-integrated learning experiences. The Blueprint distributes teaching and learning responsibilities across learners, educators, and industry/community in a three-way partnership model. Learning experiences were designed to reflect knowledge and skills relevant to professional practice. In phase two, the proof-of-concept was applied to two design units, one undergraduate and one postgraduate, and we tested acceptance and scalability, with positive student feedback. Phase three showcases how the Blueprint then informed curriculum redesign within the design school’s flagship course, including trials of ungraded assessment to further support authentic learning. The findings demonstrate that a co-created, scaffolded approach integrating industry-engaged experiences from enrolment to graduation can bridge the gap between academia and professional practice. This study contributes a practical framework for embedding authentic learning at scale, offering insights for institutions seeking to enhance employability through curriculum innovation. Full article

This study examines how system resilience can be strained when governments define and mistranslate epistemological uncertainties into technical and managerial problems, using two comparative case studies of sociotechnical disasters (i.e., nuclear safety failure and pandemic digital surveillance). Drawing on Funtowicz and Ravetz’s post-normal science as a framework, the analysis introduces three types of uncertainties and conceptualizes the “mistranslation of uncertainties,” through which this research illuminates a structure whereby epistemic diversity is marginalized, and in turn, policy legitimacy deficits in implementation can be amplified, potentially eroding sociotechnical resilience. The article contributes to the field of sociotechnical resilience by (1) visualizing uncertainty mistranslation, which leads to a legitimacy deficit, (2) illustrating how mistranslation develops from a monolithic, technocentric understanding of uncertainties, and (3) proposing a framework of “resilience for legitimacy” that seeks to embed resilience work and co-production of responses within institutional practices. This research highlights how the co-production of epistemological translation and social order, encompassing deliberation and social feedback, can support democratic legitimacy and thereby resilient governance of sociotechnical systems. Full article

Edge, fog, and cloud continuum architectures that interconnect resource-constrained devices, intermediate edge servers, and remote cloud data centers face persistent challenges in handling heterogeneous and latency-sensitive workloads while reducing energy consumption and improving resource utilization. Classical task offloading approaches either rely on static heuristics, which lack adaptability to dynamic conditions, or on metaheuristic optimizers, which often incur high computational overhead and centralized coordination. This paper proposes LITO, a lemur-inspired task offloading algorithm for edge, fog, and cloud continuum systems that models the infrastructure as a social system in which computing nodes assume distinct roles that mirror lemur social hierarchies. Building on an abstracted model of lemur group behavior, LITO incorporates two key lemur-inspired mechanisms: an energy-aware task assignment mechanism based on sun basking, a thermoregulation behavior in which lemurs seek favorable warm spots, mapped here to selecting energetically efficient execution nodes, and a cooperative scheduling policy based on huddling, group clustering under stress, mapped here to sharing load among overloaded nodes. These mechanisms are combined with a continual supervised policy-learning layer with contextual bandit feedback that refines offloading decisions from online feedback. The resulting multi-objective formulation jointly minimizes energy consumption and deadline violations while maximizing resource utilization and throughput under high-load conditions in the edge and fog segment of the continuum. Simulations under diverse workload regimes and task complexities show that LITO outperforms representative multi-objective offloading baselines in terms of energy consumption, resource utilization, latency, Service Level Agreement (SLA) violations, and throughput in congested scenarios. Full article

Despite the pervasive role of digital platforms in contemporary higher education, existing measurement tools fail to capture students’ psychological dependence on online approval within academic contexts, focusing instead on technical competencies or clinical addiction symptoms. This study developed and psychometrically validated the Digital Validation Seeking Scale (DVSS), a multidimensional instrument measuring university students’ reliance on digital feedback for academic and identity confirmation. Two independent samples of Egyptian undergraduate students were recruited: an exploratory sample of 511 students and a confirmatory sample of 740 students from six universities. The DVSS underwent rigorous content validation by eleven experts, exploratory factor analysis (EFA) using Principal Axis Factoring with Promax rotation, and confirmatory factor analysis (CFA) comparing competing structural models. Results revealed a robust four-factor structure comprising Academic Self-Quantification (ASQ), Feedback Hyper-vigilance (FHV), Social Comparison (SC), and Performative Studiousness (PS), with the first-order four-factor model demonstrating superior fit indices. The final 19-item scale exhibited excellent internal consistency, with Cronbach’s alpha coefficients ranging from 0.807 to 0.938 for subscales and total score, respectively, and strong test–retest reliability. The DVSS provides researchers and practitioners with a theoretically grounded, psychometrically sound instrument for identifying maladaptive digital validation patterns before they compromise academic engagement or psychological well-being, enabling targeted interventions within hybrid educational environments. Full article