Search Results (78)

Saturated diving plays an important role in fields such as navigation operations, ocean development, military oceanography, and maritime rescue and is indispensable for the marine economy. Heliumspeech communication is an essential component of deep-sea saturated diving operations and serves as the sole means of communication in such environments. This paper presents a heliumspeech unscrambling method for saturated diving based on phonetic multi-objective optimization using deep learning. The method consists of a heliumspeech correction network and a heliumspeech unscrambling network. First, a phonetic multi-objective optimization algorithm is used to design the correction network, which reduces the demand for large heliumspeech training datasets. Then, a saturated diving working language heliumspeech corpus is used to train the heliumspeech unscrambling network. Finally, the unscrambling network processes heliumspeech using a cognitive transfer learning algorithm. During unscrambling, the network continuously converts unscrambled heliumspeech into sample data and adds them to a supervised database, forming a closed-loop control system that dynamically adjusts the network parameters to adapt to variations in heliumspeech signals. This approach not only reduces the deep learning neural network’s reliance on large training datasets but also enhances unscrambling performance, particularly under dynamically changing diving depths. Simulation experiments demonstrate that the method effectively unscrambles heliumspeech with a low word error rate and fast convergence. Full article

In sustained Spanish–Portuguese contact zones, there is often a mismatch between observable linguistic data and speakers’ views on the degree of resulting hybridity. Terms like “Portuñol/Portunhol” are suggestive of a “third” language, but little empirical evidence has been offered in support of such outcomes, and spontaneous speech does not provide the complete envelope of variation. The present study draws on data collected in Misiones province in northeastern Argentina, where vernacular Portuguese is spoken natively in rural households, and Spanish is first acquired informally. To empirically test popular notions of “Portuñol” hybridity, an array of experimental tasks has been employed to present various monolingual and mixed configurations to bilingual participants. In the sociolinguistic free-fall environment of Misiones, the potential for the emergence of a stable hybrid is at its greatest, but the increasingly fine-grained array of experimental techniques designed to force-feed putative “Portuñol” shows that even with priming, lexical borrowing is still the only consistent contact-induced phenomenon. Full article

Accurate three-dimensional temperature and salinity initial fields are essential for regional short-term ocean forecasting, but subsurface constraints remain limited in the Western Indian Ocean because in situ profiles are sparse and satellite observations mainly describe the sea surface. This study evaluated a method to improve short-term forecasts by assimilating pretrained language model (PLM)-reconstructed thermohaline fields into the Finite Volume Community Ocean Model with a three-dimensional variational data assimilation scheme (FVCOM–3DVAR). The method was applied to the Western Indian Ocean, covering 15° S–10° N, 33° E–60° E. A non-assimilation control experiment (Control_run), Modular Ocean Data Assimilation System assimilation (MODAS_ass), and PLM reconstructed-field assimilation (PLM_ass) were conducted to evaluate the forecast performance with 5-day rolling forecasts. World Ocean Database (WOD) profiles were used for profile-based validation, and Copernicus Marine Environment Monitoring Service (CMEMS) gridded fields were used for spatially continuous reference evaluation in February, May, August, and November 2021. The results demonstrated that PLM_ass produced lower temperature and salinity root-mean-square errors (RMSEs) than Control_run and MODAS_ass at most depths and lead times. Relative to Control_run, PLM_ass reduced mean temperature RMSE by 28.0% at 100–200 m and mean salinity RMSE by 23.5% at 0–100 m; compared with MODAS_ass, the reductions were approximately 16.6% and 14.2%, respectively. Spatial diagnostics showed variable-, depth-, and region-dependent impacts with the most robust improvement in thermocline temperature forecasts. This study provides a feasible pathway for incorporating AI-reconstructed subsurface information into regional ocean forecasting systems under sparse observation conditions. Full article

The increasing reliance on digital infrastructures has positioned electronic records management systems (ERMS) as critical mechanisms for supporting organisational governance, accountability, transparency, and effective service delivery. This study presents a structured qualitative literature review examining ERMS implementation across developed and developing institutional contexts to identify key determinants, recurring implementation challenges, and contextual variations in adoption patterns. Drawing on studies published between 2012 and 2026, the review adopts a socio-technical analytical framework that categorises implementation determinants into environmental, technological, and organisational dimensions, specifically: governance and policy alignment; technological infrastructure readiness; interoperability and system integration; and human re-source capacity and organisational culture. The findings indicate that successful ERMS implementation depends on the alignment and interaction of governance frameworks, technological capabilities, and organisational readiness. The analysis further demonstrates that these determinants are highly interdependent and vary according to levels of institutional and digital maturity. In developing contexts, implementation is primarily constrained by inadequate infrastructure, financial limitations, weak policy enforcement, and shortages of skilled personnel. In contrast, digitally mature environments increasingly focus on interoperability, metadata standardisation, usability optimisation, and long-term digital preservation. This study contributes to the literature by synthesising fragmented empirical findings into an integrated socio-technical framework, thereby advancing a more structured understanding of ERMS implementation across diverse governance environments. The review also identifies major methodological limitations within the existing literature, including limited empirical validation, weak longitudinal analysis, language bias, and the predominance of single-institution case study designs. The findings provide practical implications for policymakers, information managers, and institutions seeking to strengthen electronic records management and information governance practices. Future research should prioritise longitudinal, comparative, and cross-national studies to further advance theoretical and empirical understanding of ERMS implementation. Full article

Adolescents growing up in multilingual regions experience diverse educational contexts that may shape their daily routines and psychosocial environments, but their independent relevance for mental health remains unclear. South Tyrol, with its parallel German-, Italian-, and Ladin-language school systems, provides a unique setting to examine these associations. This study assessed whether school language and home–school language mismatch are associated with mental health, psychosomatic symptoms, and health-related behaviors among adolescents. We analyzed data from a population-based survey of 2005 adolescents aged 11–19 years who provided self-reported information on mental health, psychosomatic complaints, school stress, social support, digital behaviors, lifestyle, and sleep. Multivariable regression analyses examined the independent association of home–school language mismatch with mental health outcomes, adjusting for sociodemographic and educational factors and further incorporating sleep-related behaviors. Mental health outcomes, psychosomatic symptoms, and most health-related behaviors showed little variation by school language, with generally small effect sizes. Home–school language mismatch was associated with slightly higher depressive symptom scores in unadjusted analyses but was not independently associated with mental health outcomes after adjustment. In contrast, weekly sleep problems emerged as the strongest correlate of depressive symptoms, accounting for a substantial proportion of explained variance. These findings indicate that adolescent mental health in this multilingual context is associated less with the language of schooling itself than with broader behavioral and developmental factors, highlighting sleep-related behaviors as a central and modifiable target for prevention. Full article

This systematic review investigates the application of computer-vision technologies for automated monitoring of personal protective equipment compliance in industrial environments. This review followed the PRISMA 2020 guidelines and covered studies published between 2010 and 24 February 2026. It provides a structured synthesis of advances in deep learning-based object detection models, with particular emphasis on different YOLO variants, two-stage detectors such as Faster R-CNN, and emerging transformer-based and vision–language models. Model effectiveness, reported performance metrics, and dataset characteristics are comparatively examined, including their performance under practical operating conditions. Special attention is given to performance variability in real-world scenarios affected by illumination changes, occlusion, viewing angle variation, worker movement, computational constraints, and large-scale deployment requirements. The review also appraises the reporting quality and risk of bias of the included studies and identifies current research trends, methodological limitations, and the gap between laboratory validation and industrial implementation. It also outlines future directions for improving the reliability, cost-effectiveness, and practical application of computer vision-based personal protective equipment compliance systems. Full article

Large language models (LLMs) are increasingly being used in education, yet their correctness alone does not capture the quality, reliability, or pedagogical validity of their problem-solving behavior, especially in mathematics, where multi-step logic, symbolic reasoning, and conceptual clarity are critical. Conventional evaluation methods largely focus on final answer accuracy and overlook the reasoning process. To address this gap, we introduce a novel interpretability framework for analyzing LLM-generated solutions using undergraduate calculus problems as a representative domain. Our approach combines reasoning flow extraction and decomposing solutions into semantically labeled operations and concepts with prompt ablation analysis to assess input salience and output stability. Using structured metrics such as reasoning complexity, phrase sensitivity, and robustness, we evaluated the model behavior on real Calculus I–III university exams and compared it with the performances of students enrolled in the courses. Our findings revealed that LLMs often produce syntactically fluent yet conceptually flawed solutions with reasoning patterns sensitive to prompt phrasing and input variation. This framework enables a fine-grained diagnosis of reasoning failures, supports curriculum alignment, and informs the design of interpretable AI-assisted feedback tools. The framework was evaluated on Gemma 3, an open-access large language model, across zero-shot, retrieval-augmented generation, and contextual retrieval configurations, using nine real undergraduate calculus examinations from three course levels. To our knowledge, this is the first paper to apply a combined reasoning flow decomposition and prompt ablation framework to real undergraduate calculus examinations, benchmarked against actual student cohort performance, laying the foundation for the transparent and responsible deployment of AI in STEM learning environments. Full article

Sign language understanding for automated vehicles sits at the intersection of accessibility, intelligent transportation, and safety-critical human–machine interaction. The existing sign-language recognition systems are largely confined to controlled environments, limiting their utility in mobility scenarios characterized by lighting variation, motion blur, and partial occlusion. This paper proposes STCM-HVNet, a safety-aware hybrid multimodal architecture integrating four components: a spatial visual encoder, a MediaPipe-based pose encoder, a bidirectional LSTM temporal encoder, and a context-aware fusion and safety decision module. The architecture is formulated as a multi-task system that jointly predicts sign category, interaction intent, and urgency level, and incorporates confidence-aware rejection and fail-safe action mapping. Experiments are conducted on two Arabic sign-language resources. On the RGBArS image benchmark (31 classes, 7856 images), the proposed pipeline achieves a Top-1 accuracy of 45.38%, Top-3 accuracy of 75.15%, and Macro-F1 of 0.4479, outperforming LinearECOC, kNN-5, and Bagged Trees baselines. On the Arabic sign-language video benchmark (12 classes, 479 clips), the BiLSTM temporal encoder achieves a Top-1 accuracy of 93.15% and Macro-F1 of 0.9383, outperforming frame-aggregation (87.67%) and CNN-LSTM (89.04%) baselines. Ablation results confirm complementary contributions from the visual and pose branches. A safety-threshold analysis and a Monte Carlo dropout comparison demonstrate that the proposed safety decision/gating layer provides a controllable trade-off between prediction coverage and reliability. Full article

Background: Robust perception involves processing heterogeneous sensory signals, such as facial expressions, vocal prosody, and language, particularly in noisy environments. In computational modeling, a key challenge is integrating these diverse inputs while actively filtering uninformative variations. While recent deep learning models address this integration through complex fusion architectures, they typically aggregate features without explicit filtering modules analogous to inhibitory control. In this study, we propose Multi-modal Information Disentanglement (MInD), a computational framework designed to test the hypothesis that algorithmic noise isolation facilitates robust multisensory integration. Methods: Drawing conceptual inspiration from cognitive theories of modularity, our model decomposes sensory inputs into amodal (modality-invariant) and modal-specific pathways. Furthermore, we introduce an adversarial noise isolation mechanism to serve as an algorithmic analog to cognitive inhibition. Given that our model operates on pre-extracted high-level features, this mechanism functions to isolate latent distributional variance—uninformative fluctuations that persist after initial feature extraction—guiding the network to separate task-relevant affective cues from irrelevant feature variance. Results: Empirical evaluations on standard emotion recognition benchmarks indicate that this purification-before-fusion strategy is associated with competitive performance and stability across multiple metrics. Notably, the framework attains these results using simple linear integration layers, suggesting that separating representations prior to fusion may reduce the computational complexity required for subsequent integration. Conclusions: These observations highlight the computational utility of algorithmic noise suppression, illustrating how cognitive inspiration can inform efficient machine learning architectures without claiming direct neurobiological validation. Full article

Postharvest quality of carrot (Daucus carota L.) is determined by the interaction between genotype and storage environment, yet systematic comparative evidence across pigmented genotypes with contrasting biochemical profiles remains scarce. This study evaluated the postharvest behavior of five carrot genotypes (6KUR, 14BER, yellow, white, and purple) under refrigeration (4 °C) and room temperature (15 °C) over 30 days, integrating physicochemical, spectral, and consumer-based assessment. Variables included color, fresh weight loss, respiration rate, firmness, titratable acidity, total soluble solids, and β-carotene quantification by spectrophotometry. Non-destructive monitoring was performed using Vis/NIR reflectance (350–1900 nm) with spectral indices sensitive to anthocyanin and carotenoid content (CRI1, CRI2, mARI) and tissue structural integrity (NDVI), and consumer perception (~60 participants per evaluation) was characterized through natural language processing of open-ended responses. Refrigeration significantly reduced β-carotene degradation (~15–20% loss vs. 50–60% at room temperature) and better preserved overall quality across genotypes. Purple carrots demonstrated superior postharvest stability across multiple traits, whereas white carrots showed the greatest susceptibility to quality loss. Spectral indices exhibited genotype-dependent temporal variation, particularly in pigmented roots, supporting their potential for non-destructive pigment monitoring during storage. Consumer descriptors reflected a progressive decline in desirable sensory attributes under both conditions. These findings support the integration of physicochemical, spectral, and sensory approaches for comprehensive postharvest characterization of genotypically diverse carrot germplasm, and identify priority genotypes and trait combinations for future predictive modeling studies. Full article

Genetic Programming (GP) for evolving Behavior Trees (BTs) in autonomous robots often suffer from premature convergence, even when adaptive mutation mechanisms are employed. This paper proposes a novel hybrid framework that integrates Large Language Model (LLM) supervision into GP, in which the LLM performs holistic population analysis, adaptively regulates mutation rates, and generates targeted BTs to proactively address behavioral gaps in the evolving population. Unlike conventional evolutionary operators, the LLM introduces high-level semantic guidance by seeding underrepresented behavioral archetypes, thereby complementing stochastic genetic variation with structured exploration. The proposed method is evaluated in a Unity-based multi-task robotic simulation environment. Experimental results show that the hybrid approach significantly outperforms baseline GP with standard adaptive mutation, achieving a 71.7% faster emergence of Complete Robots, a 65.2% faster emergence of Excellent Robots, and a 28% increase in behavioral diversity. Notably, the two systems exhibit opposite mutation dynamics: the LLM-guided system progressively reduces mutation rates to promote exploitation, whereas the baseline maintains a high mutation rate. In addition, the LLM generates approximately 40 targeted BTs per run, proactively seeding the population with underrepresented behavioral archetypes. These performance gains are obtained with only a 13% computational overhead. Full article

Digital social platforms have transformed political discourse into complex socio-technical environments characterized by rapid narrative diffusion, emotional amplification, and large-scale audience interaction. Understanding how sentiment and semantic alignment interact within such environments is important for analyzing polarization and patterns of audience response. This study examines narrative–audience interaction in Telegram political ecosystems using a combination of sentiment analysis, semantic similarity measures, and engagement metrics. Transformer-based language models are applied to quantify relationships between source posts and user-generated comments, enabling joint analysis of affective tone and topical alignment. The results reveal a consistent affective–semantic asymmetry: user responses tend to remain semantically aligned with source narratives while shifting toward more negative sentiment. This pattern indicates that disagreement is predominantly expressed through affective reframing rather than through divergence from the original topic. Further analysis shows systematic differences across ideological groups. Pro-government channels exhibit higher reach and more stable discourse alignment, while pro-opposition channels generate stronger engagement and more pronounced negative sentiment shifts. Neutral channels display intermediate characteristics. These findings demonstrate that online political discourse in Telegram is characterized by stable topical anchoring combined with systematic variation in emotional response. Full article

Objective: Functional training exercises involve complex multi-joint movements that challenge traditional rule-based or data-driven recognition systems. This paper introduces a Movement Description Language (MDL) designed to formally represent, analyze, and evaluate such exercises using camera-based pose estimation and interpretable, composable structures. Methods: The proposed MDL models each exercise as a finite-state machine defined by pose-derived angle proxy transitions, allowing movements to be described in a modular and reusable way. Demonstrated with MediaPipe landmark extraction from monocular video, while the MDL remains compatible with any pose estimation algorithm, the framework focuses on exercise phase detection and repetition counting. Experimental validation was conducted on a dataset of 1513 videos of 12 functional exercises (squats, deadlifts, lunges, shoulder presses, planks, push-ups, pull-ups, bent-over rows, box jumps, thrusters, overhead squats, and burpees) obtained from public pose datasets, competition footage, and recordings of 9 participants in real-world environments. Results: Automated repetition counts were compared against manually annotated ground truth, showing an overall repetition-counting accuracy of 97.2%, with a mean per-exercise accuracy of 98.8% (range 95–100%). The MDL successfully handled both simple and compound exercises, maintaining reliable phase detection despite variations in execution speed, camera perspective, and environmental conditions. Conclusions: The system was implemented using real-time pose estimation to demonstrate the practical execution of the MDL framework. The proposed MDL provides a transparent, extensible, and computationally efficient framework for functional exercise analysis. By bridging human-readable movement semantics with executable motion logic, it enables interpretable automatic repetition counting and phase detection, offering an alternative to black-box recognition approaches. The results support its potential for scalable deployment in training, monitoring and movement analysis applications. The proposed system is not intended for biomechanical measurement or clinical-grade kinematic analysis, but rather for interpretable modeling of exercise structure and repetition detection using approximate pose-derived signals. Full article

Dialectal variation presents a major challenge for deploying medical language models in real-world healthcare settings, where patient–clinician communication often occurs in regional vernaculars rather than standardized language forms. This challenge is particularly pronounced in the Arabic-speaking world, where clinical interactions frequently take place in diverse dialects that differ substantially from Modern Standard Arabic. Fine-tuning and maintaining separate models for each dialect is computationally inefficient and difficult to scale, motivating more integrated approaches. In this work, we present MENARA, an Arabic medical language model constructed by merging Egyptian Arabic, Moroccan Darija, and medical-domain specialists through model merging. We extend prior feasibility findings through comprehensive evaluation of cross-dialect performance, medical safety, and cross-lingual knowledge retention. Specifically, we introduce a fine-grained dialect composition analysis to quantify lexical purity and structured code-switching behavior, benchmark against state-of-the-art Arabic LLMs, conduct subject-matter-expert assessment of both dialectal fidelity and medical appropriateness. The results show that model merging preserves core medical competence while enabling robust dialectal adaptation, achieving strong cross-dialect fidelity while substantially reducing storage and deployment overhead compared to maintaining separate models. These findings establish model merging as a potentially practical and resource-efficient paradigm for dialect-aware medical NLP in linguistically fragmented healthcare environments. Full article

Syntactic complexity is a crucial aspect of assessing writing quality in Chinese as a second language. While existing literature predominantly focuses on synchronic features of syntactic complexity, particularly changes in complexity indices, less attention has been paid to its diachronic development and interactions among these indices. Drawing upon Dynamic Systems Theory, this exploratory longitudinal study traces the developmental trajectories of syntactic complexity indices and their interactions in CSL writings of 15 native Cambodian speakers within a single instructional context. The main results are as follows: (i) The syntactic complexity indices exhibited fluctuating and nonlinear growth. Among them, the length of topic chain clauses showed notable variation (range: 1.12 to 9.05), while relatively small changes (range: 0.4 to 2.39) occurred in the number of topic chain clauses. (ii) The development trends in the number of topic chains and the number of zero components showed no significant difference (p = 0.086, Cohen’s d = 0.31). In contrast, the development trends in the number of topic chain clauses and the length of topic chain clauses differed significantly (p = 0.039, Cohen’s d = 0.65). (iii) Individual differences in syntactic complexity were observed among learners in similar learning environments. These findings provide a detailed, context-bound description of the dynamic and complex syntactic development observed in the Chinese writing of 15 participants. The study highlights the value of employing multiple perspectives to capture such complexity and underscores the need for future research with more diverse samples and designs to test the generalizability of these patterns. Full article