Search Results (1,090)

Recent advancements in large language models (LLMs) have significantly enhanced the effectiveness of Retrieval-Augmented Generation (RAG) systems. This study focuses on the development and evaluation of a domain-specific AI chatbot designed to support international student admissions by leveraging LLM-based RAG pipelines. We implement and compare multiple pipeline configurations, combining retrieval methods (e.g., Dense, MMR, Hybrid), chunking strategies (e.g., Semantic, Recursive), and both open-source and commercial LLMs. Dual evaluation datasets of LLM-generated and human-tagged QA sets are used to measure answer relevancy, faithfulness, context precision, and recall, alongside heuristic NLP metrics. Furthermore, latency analysis across different RAG stages is conducted to assess deployment feasibility in real-world educational environments. Results show that well-optimized open-source RAG pipelines can offer comparable performance to GPT-4o while maintaining scalability and cost-efficiency. These findings suggest that the proposed chatbot system can provide a practical and technically sound solution for international student services in resource-constrained academic institutions. Full article

Background/Objectives: Despite legal frameworks acknowledging the need to protect the rights of female prisoners, penitentiary systems often neglect gender-specific needs, particularly for foreign women. Among them, Muslim women face distinct challenges linked to cultural and religious practices, which are frequently unmet in prison contexts. This review aims to explore the academic literature on the experiences of Muslim women in detention. Methods: A systematic review was conducted using three major bibliographic databases—Scopus, PubMed, and Web of Science—covering the period from 2010 to 2024. Inclusion criteria focused on peer-reviewed studies examining the condition of Muslim women in prison. Of the initial pool, only four articles met the criteria and were included in the final analysis. Results: The review reveals a marked scarcity of research on Muslim women in prison at both national and international levels. This gap may be due to their limited representation or cultural factors that hinder open discourse. The selected studies highlight key issues, including restricted access to services, limited ability to practice religion, and language and cultural barriers. These challenges contribute to increased psychological vulnerability, which is often underestimated in prison settings. Conclusions: There is an urgent need for targeted research and culturally competent training for prison staff to adequately support Muslim women in detention. Greater academic and institutional attention is essential to develop inclusive policies that consider the intersection of gender, religion, and migration, particularly in the post-release reintegration process. Full article

Academic English support is crucial for English as an Additional Language (EAL) nursing students in English-medium nursing education programmes. However, empirical research on content and language integrated learning (CLIL) within this specific context remains limited. This study, informed by recent advancements in translanguaging and trans-semiotising (TL-TS) theory, investigates the patterns of teacher collaboration in nursing CLIL and its impact when employing a TL-TS pedagogical approach. Analysis of students’ pre- and post-tests and multimodal classroom interactions reveals that effective collaboration between nursing specialists and language experts in CLIL can be fostered by (1) aligning with language education principles through the incorporation of internally persuasive discourse (IPD) about language learning and TL-TS practices; (2) simulating potential professional contingencies and co-developing coping strategies using TL-TS; and (3) elucidating nursing language norms through TL-TS and IPD. We advocate for re-imagination of CLIL in English-medium nursing education through an organistic–procedural TL perspective and highlight its potential to enhance EAL nursing students’ development of language proficiency and professional competencies. Full article

This study analyzes the relationship between ergonomic and psychosocial risk factors and labor productivity using a bibliometric approach through a general analysis and one that includes inclusion criteria such as English language, open access, and primary research publications to identify only those articles that explicitly address the relationship between ergonomic and psychosocial risk factors and labor productivity. It is recognized that both physical and psychosocial conditions of the work environment directly influence workers’ health and organizational performance. For this purpose, a bibliometric review was conducted in academic databases, including Scopus, Web of Science, ScienceDirect, and Taylor & Francis, resulting in the selection of 4794 relevant articles for general analysis. Additionally, 116 relevant articles were selected based on the inclusion criteria. Tools and methodologies, such as Rayyan, Excel, VOSviewer 1.6.20, and PRISMA, were used to classify the studies and identify trends, collaboration networks, and geographical distribution. The results reveal a sustained growth in scientific production, with clusters on occupational safety and health, work environment factors, and the characteristics of the population, approach, and methodologies used in the studies. Likewise, Procedia Manufacturing, International Journal of Occupational Safety and Ergonomics, and Ergonomics stand out as the main sources of publication, while countries such as Sweden, Poland, and the United States lead the scientific production in this field. In addition, the network of co-occurrence of keywords evidences a comprehensive approach that articulates physical or ergonomic and psychosocial risk factors with organizational performance, while the network of authors shows consolidated collaborations and studies focused on analyzing the relationship between physical demands and musculoskeletal disorders from advanced ergonomic approaches. Full article

(1) Background and objectives: Large language models (LLMs) such as GPT, Mistral, and LLaMA exhibit strong capabilities in text generation, yet assessing the quality of their reasoning—particularly in open-ended and argumentative contexts—remains a persistent challenge. This study introduces Dialectical Agent, an internally developed modular framework designed to evaluate reasoning through a structured three-stage process: opinion, counterargument, and synthesis. The framework enables transparent and comparative analysis of how different LLMs handle dialectical reasoning. (2) Methods: Each stage is executed by a single model, and final syntheses are scored via two independent LLM evaluators (LLaMA 3.1 and GPT-4o) based on a rubric with four dimensions: clarity, coherence, originality, and dialecticality. In parallel, a rule-based semantic analyzer detects rhetorical anomalies and ethical values. All outputs and metadata are stored in a Neo4j graph database for structured exploration. (3) Results: The system was applied to four open-weight models (Gemma 7B, Mistral 7B, Dolphin-Mistral, Zephyr 7B) across ten open-ended prompts on ethical, political, and technological topics. The results show consistent stylistic and semantic variation across models, with moderate inter-rater agreement. Semantic diagnostics revealed differences in value expression and rhetorical flaws not captured by rubric scores. (4) Originality: The framework is, to our knowledge, the first to integrate multi-stage reasoning, rubric-based and semantic evaluation, and graph-based storage into a single system. It enables replicable, interpretable, and multidimensional assessment of generative reasoning—supporting researchers, developers, and educators working with LLMs in high-stakes contexts. Full article

This study synthesizes the rapidly expanding evidence on how digital technologies reshape international trade, with a particular focus on small and medium-sized enterprises (SMEs). Guided by two research questions—(RQ1) How do digital tools influence the volume and composition of cross-border trade? and (RQ2) How do these effects vary by countries’ development level and firm size?—we conducted a PRISMA-compliant systematic literature review covering 2010–2024. Searches across eight major databases yielded 1857 records; after duplicate removal, title/abstract screening, full-text assessment, and Mixed Methods Appraisal Tool (MMAT 2018) quality checks, 86 peer-reviewed English-language studies were retained. Findings reveal three dominant technology clusters: (1) e-commerce platforms and cloud services, (2) IoT-enabled supply chain solutions, and (3) emerging AI analytics. E-commerce and cloud adoption consistently raise export intensity—doubling it for digitally mature SMEs—while AI applications are the fastest-growing research strand, particularly in East Asia and Northern Europe. However, benefits are uneven: firms in low-infrastructure settings face higher fixed digital costs, and cybersecurity and regulatory fragmentation remain pervasive obstacles. By integrating trade economics with development and SME internationalization studies, this review offers the first holistic framework that links national digital infrastructure and policy support to firm-level export performance. It shows that the trade-enhancing effects of digitalization are contingent on robust broadband penetration, affordable cloud access, and harmonized data-governance regimes. Policymakers should, therefore, prioritize inclusive digital-readiness programs, while business leaders should invest in complementary capabilities—data analytics, cyber-risk management, and cross-border e-logistics—to fully capture digital trade gains. This balanced perspective advances theory and practice on building resilient, equitable digital trade ecosystems. Full article

This study presents a lightweight autoencoder-based approach for anomaly detection in digit recognition using federated learning on resource-constrained embedded devices. We implement and evaluate compact autoencoder models on the ESP32-CAM microcontroller, enabling both training and inference directly on the device using 32-bit floating-point arithmetic. The system is trained on a reduced MNIST dataset (1000 resized samples) and evaluated using EMNIST and MNIST-C for anomaly detection. Seven fully connected autoencoder architectures are first evaluated on a PC to explore the impact of model size and batch size on training time and anomaly detection performance. Selected models are then re-implemented in the C programming language and deployed on a single ESP32 device, achieving training times as short as 12 min, inference latency as low as 9 ms, and F1 scores of up to 0.87. Autoencoders are further tested on ten devices in a real-world federated learning experiment using Wi-Fi. We explore non-IID and IID data distribution scenarios: (1) digit-specialized devices and (2) partitioned datasets with varying content and anomaly types. The results show that small unmodified autoencoder models can be effectively trained and evaluated directly on low-power hardware. The best models achieve F1 scores of up to 0.87 in the standard IID setting and 0.86 in the extreme non-IID setting. Despite some clients being trained on corrupted datasets, federated aggregation proves resilient, maintaining high overall performance. The resource analysis shows that more than half of the models and all the training-related allocations fit entirely in internal RAM. These findings confirm the feasibility of local float32 training and collaborative anomaly detection on low-cost hardware, supporting scalable and privacy-preserving edge intelligence. Full article

This paper aims to discuss the transition of new graduate nurses into the workforce, the preparation provided to equip them through the novice–beginner stage, and the theory–practice conundrum. Background: In Australia, new graduate transition programs have been in existence since the 1990s. While there is widespread acknowledgment that this period is pivotal for new graduate nurses entering the profession, there is a lack of consensus on the definition of best practice to achieve optimal preparation for new graduate nurses transitioning into the workforce. Methods: This discussion paper integrates the nursing literature on this topic with the extensive professional experiences of the authors, who are currently working as clinicians in metropolitan hospitals and hold academic positions at universities. Their insights are informed by the literature sourced from peer-reviewed English language journals, including reviews, empirical studies, and national and international reports. Discussion: Recruiting and retaining nurses presents a multifaceted challenge that requires the development of effective tools and strategies to build a sustainable workforce. Both the literature and the authors’ experiences highlight several key factors influencing the preparedness of new graduates. These factors include workplace culture, the demands placed on new graduates, and the support, education, and training they receive. The perspectives shared in this article offer valuable discussion points that can deepen our understanding of the current issues and contribute to the development of more effective solutions. Full article

Background/Objectives: Art therapy is a psychotherapeutic technique that involves the creation of tangible visual arts and represents a coping strategy to support children with cancer. Evaluating the effects of such activities on children with cancer is essential for providing evidence of the value that creativity holds within healthcare systems. A dedicated tool for assessing the creative process is the Arts Observational Scale (ArtsObS), focusing on mood and emotional states as key indicators of psychosocial well-being. This study aims to validate a translated version of the ArtsObS in the Italian language. Methods: The translation process followed recommendations for translation and cultural adaptation. The distribution properties of the scores, internal consistency, sensitivity to change, reliability, and convergent validity were assessed through observations conducted by two different evaluators. Results: The ArtsObS in its Italian adaptation is proven to be an adequate tool for capturing changes following an intervention, with good internal consistency and low sensitivity to differences between operators. The analysis supports the reliability of the ArtsObS across different observers. Conclusions: The Italian ArtsObS is a valid and reliable instrument for evaluating the impact of art therapy on pediatric patients’ mood and emotional states. It provides a standardized tool for clinical and research settings to assess creative interventions in pediatric oncology. Full article

Background/Objectives: The demand for accessible and reliable digital health services has increased significantly in recent years, particularly in regions facing physician shortages. HomeDOCtor, a conversational AI platform developed in Slovenia, addresses this need with a nationally adapted architecture that combines retrieval-augmented generation (RAG) and a Redis-based vector database of curated medical guidelines. The objective of this study was to assess the performance and impact of HomeDOCtor in providing AI-powered healthcare assistance. Methods: HomeDOCtor is designed for human-centered communication and clinical relevance, supporting multilingual and multimedia citizen inputs while being available 24/7. It was tested using a set of 100 international clinical vignettes and 150 internal medicine exam questions from the University of Ljubljana to validate its clinical performance. Results: During its six-month nationwide deployment, HomeDOCtor received overwhelmingly positive user feedback with minimal criticism, and exceeded initial expectations, especially in light of widespread media narratives warning about the risks of AI. HomeDOCtor autonomously delivered localized, evidence-based guidance, including self-care instructions and referral suggestions, with average response times under three seconds. On international benchmarks, the system achieved ≥95% Top-1 diagnostic accuracy, comparable to leading medical AI platforms, and significantly outperformed stand-alone ChatGPT-4o in the national context (90.7% vs. 80.7%, p = 0.0135). Conclusions: Practically, HomeDOCtor eases the burden on healthcare professionals by providing citizens with 24/7 autonomous, personalized triage and self-care guidance for less complex medical issues, ensuring that these cases are self-managed efficiently. The system also identifies more serious cases that might otherwise be neglected, directing them to professionals for appropriate care. Theoretically, HomeDOCtor demonstrates that domain-specific, nationally adapted large language models can outperform general-purpose models. Methodologically, it offers a framework for integrating GDPR-compliant AI solutions in healthcare. These findings emphasize the value of localization in conversational AI and telemedicine solutions across diverse national contexts. Full article

This paper examines the philosophy of language in The Platform Sutra of the Sixth Patriarch (六祖壇經), demonstrating its centrality to Zen Buddhism and Buddhist sinicization. The sutra emphasizes the ineffability of ultimate truth (至道無言) and the principle that words falter in encapsulating the Dao (言語道斷), framing language as a provisional “raft” (筏) that must be instrumentalized yet transcended through a dialectic of employing and abandoning (用離辯證). It ontologically grounds this view in Buddha-nature’s (佛性) pre-linguistic essence, advocating transcending reliance on words and letters (不假文字) while strategically deploying language to dismantle its own authority. Historically, this constituted a revolt against Tang scholasticism’s textual fetishism. The text adopts a dynamic dialectic, neither clinging to nor rejecting language, exemplified by Huineng’s awakening through the Diamond Sutra, where recitation catalyzes internal insight. Operationally, it utilizes negational discourse, the “Two Paths Mutually Condition” method (二道相因) embedded in the “Twelve Pairs of Dharmic Forms” (法相語言十二對) in particular, to systematically deconstruct dualisms, while promoting embodied unity of speech, mind, and action (口念心行) to critique empty recitation. Ultimately, the sutra orchestrates language as a self-subverting medium: balancing acknowledgment of its limitations with pragmatic instrumentality, it presents an Eastern paradigm where language actively disrupts conceptual fetters to facilitate direct insight into Buddha-nature, reframing it as a dynamic catalyst for “illuminating the mind and seeing one’s nature” (明心見性). Full article

This study examines the geostrategic role of Hungarian-language higher education institutions beyond Hungary’s border. These institutions not only fulfil an educational function but also play a role in preserving identity and geopolitics in the national policy of the Hungarian state. This research is based on a narrative review of the literature, which analyses the demographic situation of Hungarians living beyond the borders and the tools used to support higher education by synthesising domestic and international literature, statistical data, and forecasts. The results highlight that Hungarian-language higher education plays a key role in preserving ethnocultural identity and increasing the chances of success in the homeland, but also faces constraints such as labour market disadvantages resulting from a lack of state language skills. This study concludes that, in order to ensure the sustainability of Hungarian higher education beyond the border, it is necessary to strike a balance between identity preservation and integration, thereby promoting geopolitical stability and cultural cohesion with the majority society. Full article

This study investigates how differing national research governance models impact academic performance by comparing European Research Council (ERC) grant winners in the social sciences and humanities from France and the Netherlands. Situated within the broader context of centralized versus decentralized research systems, the analysis aims to understand how these structures shape publication trends, thematic diversity, and collaboration patterns. Drawing on Scopus and SciVal data covering 9996 publications by 305 ERC winners between 2019 and 2023, we employed a multi-method approach, including latent Dirichlet allocation for topic modeling, compound annual growth rate analysis, and co-authorship network analysis. The results show that neuroscience, climate change, and psychology are dominant domains, with language and linguistics particularly prevalent in France and law and political science in the Netherlands. French ERC winners are more likely to be affiliated with national or sectoral institutions, whereas in the Netherlands, elite universities dominate. Collaboration emerged as a key success factor, with an average of four co-authors per publication and network analyses revealing central figures who bridge topical clusters. International collaborations were consistently linked with higher visibility, while single-authored publications showed limited impact. These findings suggest that institutional context and collaborative practices significantly shape research performance in both countries. Full article

The rapid advancement of artificial intelligence (AI) and machine learning has revolutionised how systems process information, make decisions, and adapt to dynamic environments. AI-driven approaches have significantly enhanced efficiency and problem-solving capabilities across various domains, from automated decision-making to knowledge representation and predictive modelling. These developments have led to the emergence of increasingly sophisticated models capable of learning patterns, reasoning over complex data structures, and generalising across tasks. As AI systems become more deeply integrated into networked infrastructures and the Internet of Things (IoT), their ability to process and interpret data in real-time is essential for optimising intelligent communication networks, distributed decision making, and autonomous IoT systems. However, despite these achievements, the internal mechanisms that drive LLMs’ reasoning and generalisation capabilities remain largely unexplored. This lack of transparency, compounded by challenges such as hallucinations, adversarial perturbations, and misaligned human expectations, raises concerns about their safe and beneficial deployment. Understanding the underlying principles governing AI models is crucial for their integration into intelligent network systems, automated decision-making processes, and secure digital infrastructures. This paper provides a comprehensive analysis of explainability approaches aimed at uncovering the fundamental mechanisms of LLMs. We investigate the strategic components contributing to their generalisation abilities, focusing on methods to quantify acquired knowledge and assess its representation within model parameters. Specifically, we examine mechanistic interpretability, probing techniques, and representation engineering as tools to decipher how knowledge is structured, encoded, and retrieved in AI systems. Furthermore, by adopting a mechanistic perspective, we analyse emergent phenomena within training dynamics, particularly memorisation and generalisation, which also play a crucial role in broader AI-driven systems, including adaptive network intelligence, edge computing, and real-time decision-making architectures. Understanding these principles is crucial for bridging the gap between black-box AI models and practical, explainable AI applications, thereby ensuring trust, robustness, and efficiency in language-based and general AI systems. Full article

This systematic review aimed to evaluate the effectiveness of teriparatide (TP) in guided bone regeneration (GBR). An electronic search without language or date restrictions was performed in PubMed, Web of Science, Scopus, Scielo, and gray literature for articles published until June 2025. Inclusion criteria considered studies evaluating the effect of TP on bone regeneration, analyzed using SYRCLE’s Risk of Bias tool. Twenty-four preclinical studies were included, covering diverse craniofacial models (mandibular, calvarial, extraction sockets, sinus augmentation, distraction osteogenesis, segmental defects) and employing systemic or local TP administration. Teriparatide consistently enhanced osteogenesis, graft integration, angiogenesis, and mineralization, with potentiated effects when combined with various biomaterials, including polyethylene glycol (PEG), hydroxyapatite/tricalcium phosphate (HA/TCP), biphasic calcium phosphate (BCP), octacalcium phosphate collagen (OCP/Col), enamel matrix derivatives (EMDs), autografts, allografts, xenografts (Bio-Oss), strontium ranelate, and bioactive glass. Critically, most studies presented a moderate-to-high risk of bias, with insufficient randomization, allocation concealment, and blinding, which limited the internal validity of the findings. TP shows promising osteoanabolic potential in guided bone regeneration, enhancing bone formation, angiogenesis, and scaffold integration across preclinical models. Nonetheless, its translation to clinical practice requires well-designed human randomized controlled trials to define optimal dosing strategies, long-term safety, and its role in oral and craniomaxillofacial surgical applications. Full article