Search Results (136)

As severe selection and declining population numbers in many breeds have resulted in losses in the worldwide livestock genetic biodiversity, human concern about the situation of genetic variety in livestock breeds and their conservation has grown. In this context, genomic techniques now allow for more exact monitoring of adaptive traits and inbreeding, while reproductive techniques such as somatic cell nuclear transfer and IVF (In Vitro Fertilization) help to preserve and recover rare genetic lines. AI-powered (Artifficial Inteligence) risk assessment models and digital herdbooks contribute to data-driven reproductive strategies, particularly in smallholder settings. Nonetheless, these advances face persistent hurdles, including a lack of legislative frameworks, high costs, limited accessibility in low-resource settings, and unresolved ethical problems. The findings highlight the importance of a balanced, interdisciplinary strategy that combines new biotechnologies with traditional knowledge, collaborative practices, and strong policy to assist in preserving the long-term viability of livestock genetic resources. This review intends to assess modern and traditional methods for the preservation of livestock breeds, analyzing references published between 2019 and the present. Full article

Traditional educational assessment systems suffer from inefficient question selection strategies that fail to optimally probe student knowledge while requiring extensive testing time. We present a novel hierarchical probabilistic neural framework that integrates Bayesian inference with symmetric deep neural architectures to enable adaptive, efficient knowledge assessment. Our method models student knowledge as latent representations within a graph-structured concept dependency network, where probabilistic mastery states, updated through variational inference, are encoded by symmetric graph properties and symmetric concept representations that preserve structural equivalences across similar knowledge configurations. The system employs a symmetric dual-network architecture: a concept embedding network that learns scale-invariant hierarchical knowledge representations from assessment data and a question selection network that optimizes symmetric information gain through deep reinforcement learning with symmetric reward structures. We introduce a novel uncertainty-aware objective function that leverages symmetric uncertainty measures to balance exploration of uncertain knowledge regions with exploitation of informative question patterns. The hierarchical structure captures both fine-grained concept mastery and broader domain understanding through multi-scale graph convolutions that preserve local graph symmetries and global structural invariances. Our symmetric information-theoretic method ensures balanced assessment strategies that maintain diagnostic equivalence across isomorphic concept subgraphs. Experimental validation on large-scale educational datasets demonstrates that our method achieves 76.3% diagnostic accuracy while reducing the question count by 35.1% compared to traditional assessments. The learned concept embeddings reveal interpretable knowledge structures with symmetric dependency patterns that align with pedagogical theory. Our work generalizes across domains and student populations through symmetric transfer learning mechanisms, providing a principled framework for intelligent tutoring systems and adaptive testing platforms. The integration of probabilistic reasoning with symmetric neural pattern recognition offers a robust solution to the fundamental trade-off between assessment efficiency and diagnostic precision in educational technology. Full article

Multi-objective Unmanned Aerial Vehicle (UAV) path planning in complex 3D environments presents a fundamental challenge requiring the simultaneous optimization of conflicting objectives such as path length, safety, altitude constraints, and smoothness. This study proposes a novel hybrid framework, termed QL-MOPSO, that integrates reinforcement learning with metaheuristic optimization through a three-stage hierarchical architecture. The framework employs Q-learning to generate a global guidance path in a discretized 2D grid environment using an eight-directional symmetric action space that embodies rotational symmetry at π/4 intervals, ensuring uniform exploration capabilities and unbiased path planning. A crucial intermediate stage transforms the discrete 2D path into a 3D initial trajectory, bridging the gap between discrete learning and continuous optimization domains. The MOPSO algorithm then performs fine-grained refinement in continuous 3D space, guided by a novel Q-learning path deviation objective that ensures continuous knowledge transfer throughout the optimization process. Experimental results demonstrate that the symmetric action space design yields 20.6% shorter paths compared to asymmetric alternatives, while the complete QL-MOPSO framework achieves 5% path length reduction and significantly faster convergence compared to standard MOPSO. The proposed method successfully generates Pareto-optimal solutions that balance multiple objectives while leveraging the symmetry-aware guidance mechanism to avoid local optima and accelerate convergence, offering a robust solution for complex multi-objective UAV path planning problems. Full article

Surrogate modelling is increasingly used in engineering to improve computational efficiency in complex simulations. However, traditional data-driven surrogate models often face limitations in generalizability, physical consistency, and extrapolation—issues that are especially critical in safety-sensitive fields such as fire safety engineering (FSE). To address these concerns, physics-informed surrogate modelling (PISM) integrates physical laws into machine learning models, enhancing their accuracy, robustness, and interpretability. This systematic review synthesises existing applications of PISM in FSE, classifies the strategies used to embed physical knowledge, and outlines key research challenges. A comprehensive search was conducted across Google Scholar, ResearchGate, ScienceDirect, and arXiv up to May 2025, supported by backward and forward snowballing. Studies were screened against predefined criteria, and relevant data were analysed through narrative synthesis. A total of 100 studies were included, covering five core FSE domains: fire dynamics, wildfire behaviour, structural fire engineering, material response, and heat transfer. Four main strategies for embedding physics into machine learning were identified: feature engineering techniques (FETs), loss-constrained techniques (LCTs), architecture-constrained techniques (ACTs), and offline-constrained techniques (OCTs). While LCT and ACT offer strict enforcement of physical laws, hybrid approaches combining multiple strategies often produce better results. A stepwise framework is proposed to guide the development of PISM in FSE, aiming to balance computational efficiency with physical realism. Common challenges include handling nonlinear behaviour, improving data efficiency, quantifying uncertainty, and supporting multi-physics integration. Still, PISM shows strong potential to improve the reliability and transparency of machine learning in fire safety applications. Full article

In wind tunnel tests for aerospace and bridge engineering, the accurate prediction of Mach number remains a core challenge to ensure the reliability of airflow dynamics characterization. Pure data-driven models often fail to meet high-precision prediction requirements due to the lack of physical mechanism constraints and insufficient generalization capability. This paper proposes a physical information-based long short-term memory network (P-LSTM), which constructs a physical loss function by embedding isentropic flow equations from gas dynamics, thereby constraining the Mach number prediction solution space within the physically feasible domain. This approach effectively balances the neural network’s ability to capture temporal features with the interpretability of physical mechanisms. Aiming at the scarcity of data in new wind tunnel scenarios, an adaptive weight transfer learning method (AWTL) is further proposed, realizing efficient knowledge transfer across different-scale wind tunnels via cross-domain data calibration, adaptive source-domain weight reweighting, and target-domain fine-tuning. Experimental results show that the P-LSTM method achieves a 50.65–62.54% reduction in RMSE, 48.00–54.05% in MAE, and 47.88–73.68% in MD compared with traditional LSTM for Mach number prediction in the 0.6 m continuous wind tunnel flow field. The AWTL model also outperforms the direct training model significantly in the 2.4 m continuous wind tunnel, with RMSE, MAE, and MD reduced by 85.26%, 95.12%, and 71.14%, respectively. These results validate that the proposed models achieve high-precision Mach number prediction with strong generalization capability. Full article

This study develops and validates an artificial intelligence (AI)-assisted internship learning platform for automotive electronics based on the Llama 3 large language model, aiming to enhance pedagogical effectiveness within vocational training contexts. Addressing critical issues such as the persistent theory–practice gap and limited innovation capability prevalent in existing curricula, we leverage the natural language processing (NLP) capabilities of Llama 3 through fine-tuning based on transfer learning to establish a specialized knowledge base encompassing fundamental circuit principles and fault diagnosis protocols. The implementation employs the Hugging Face Transformers library with optimized hyperparameters, including a learning rate of 5 × 10⁻⁵ across five training epochs. Post-training evaluations revealed an accuracy of 89.7% on validation tasks (representing a 12.4% improvement over the baseline model), a semantic comprehension precision of 92.3% in technical question-and-answer assessments, a mathematical computation accuracy of 78.4% (highlighting this as a current limitation), and a latency of 6.3 s under peak operational workloads (indicating a system bottleneck). Although direct trials involving students were deliberately avoided, the platform’s technical feasibility was validated through multidimensional benchmarking against established models (BERT-base and GPT-2), confirming superior domain adaptability (F1 = 0.87) and enhanced error tolerance (σ² = 1.2). Notable limitations emerged in numerical reasoning tasks (Cohen’s d = 1.15 compared to human experts) and in real-time responsiveness deterioration when exceeding 50 concurrent users. The study concludes that Llama 3 demonstrates considerable promise for automotive electronics skills development. Proposed future enhancements include integrating symbolic AI modules to improve computational reliability, implementing Kubernetes-based load balancing to ensure latency below 2 s at scale, and conducting longitudinal pedagogical validation studies with trainees. This research provides a robust technical foundation for AI-driven vocational education, especially suited to mechatronics fields that require close integration between theoretical knowledge and practical troubleshooting skills. Full article

Automated defect detection for printed fabric manufacturing faces critical challenges in balancing industrial-grade accuracy with real-time deployment efficiency. To address this, we propose RT-DETR-FFD, a knowledge-distilled detector optimized for printed fabric defect inspection. Firstly, the student model integrates a Fourier cross-stage mixer (FCSM). This module disentangles defect features from periodic textile backgrounds through spectral decoupling. Secondly, we introduce FuseFlow-Net to enable dynamic multi-scale interaction, thereby enhancing discriminative feature representation. Additionally, a learnable positional encoding (LPE) module transcends rigid geometric constraints, strengthening contextual awareness. Furthermore, we design a dynamic correlation-guided loss (DCGLoss) for distillation optimization. Our loss leverages masked frequency-channel alignment and cross-domain fusion mechanisms to streamline knowledge transfer. Experiments demonstrate that the distilled model achieves an mAP@0.5 of 82.1%, surpassing the baseline RT-DETR-R18 by 6.3% while reducing parameters by 11.7%. This work establishes an effective paradigm for deploying high-precision defect detectors in resource-constrained industrial scenarios, advancing real-time quality control in textile manufacturing. Full article

This paper aims to explore the challenges of maintaining and modernizing legacy systems, particularly COBOL-based platforms, the backbone of many financial and administrative systems. By exploring the DOGE team’s initiative to modernize government IT systems on a relevant case study, the author analyzes the pros and cons of AI and Agile methodologies in addressing the limitations of static and highly resilient legacy architectures. A systematic literature review was conducted to assess the state of the art about legacy system modernization, AI integration, and Agile methodologies. Then, the gray literature was analyzed to provide practical insights into how government agencies can modernize their IT infrastructures while addressing the growing shortage of COBOL experts. Findings suggest that AI may support interoperability, automation, and knowledge abstraction, but also introduce new risks related to cybersecurity, workforce disruption, and knowledge retention. Furthermore, the transition from Waterfall to Agile approaches poses significant epistemological and operational challenges. The results highlight the importance of adopting a hybrid human–AI model and structured governance strategies to ensure sustainable and secure system evolution. This study offers valuable insights for organizations that are facing the challenge of balancing the desire for modernization with the need to ensure their systems remain functional and manage tacit knowledge transfer. Full article

Scandinavian erratic boulders in central Poland represent a significant element of the region’s geodiversity, fulfilling important natural, scientific, and cultural functions. As objects of high perceptual value, they integrate into the landscape and provide a wide range of geosystem services. The main objectives of research conducted in two areas of the Małopolska Upland are to determine the concentration of these boulders and identify the geosystem benefits they offer, with particular emphasis on lichen species inhabiting their surfaces. Research has confirmed the currently limited use of geosystem services provided by the 25 erratic boulders studied. However, this may change with growing ecological awareness among local communities, enabling a deeper appreciation of inanimate nature. Erratic boulders have the potential to attract geotourists and thus support economic development (by improving the residents’ quality of life), but this potential requires broader promotion. Although the Central Register of Geosites of Poland is an appropriate platform for their registration, none of the analysed boulders have yet been included. The research findings are also partly directed at local government units to help them recognise the value of erratic boulders for sustainable development, in line with existing legal frameworks and development strategies. The detailed characterisation of 25 boulders may inspire broader initiatives and foster knowledge transfer to support regional development through geotourism. The ability to identify the ecosystem benefits provided by erratic boulders is essential for maintaining ecological balance and sustaining natural processes. However, there is growing evidence of the systematic disappearance of erratic boulders from the landscape, which disrupts geosystem balance and leads to further environmental degradation, negatively affecting human well-being. In light of the lack of effective nature protection measures in the study area, it is proposed that some of these boulders be designated as geological protected features. Such a conservation approach could help maintain ecological balance in the designated area. Full article

This article explores the integration of “Ayatutu”, a communal philosophy from Nigeria’s Tiv people, into mathematics education frameworks. Ayatutu—embodying collective responsibility and mutual assistance—aligns with contemporary cooperative learning approaches while offering unique cultural dimensions. Through analysis of the ethnomathematics literature, indigenous knowledge systems, and cooperative learning theories this article develops a theoretical framework for Ayatutu-based mathematics instruction built on the following five core elements: collective problem-solving, resource sharing, complementary expertise, process orientation, and intergenerational knowledge transfer. The framework demonstrates significant alignment with sociocultural learning theory, communities of practice, and critical pedagogy while also offering potential benefits including enhanced mathematical engagement, positive identity development, stronger learning communities, and cultural sustainability. Implementation challenges involving teacher preparation, structural constraints, cultural translation, and balancing individual with collective learning are examined. This research contributes to decolonizing mathematics education by positioning indigenous philosophical systems as valuable resources for creating culturally responsive and mathematically powerful learning environments that serve diverse student populations while honoring cultural wisdom. Full article

Clinical text classification presents significant challenges in healthcare informatics due to inherent asymmetries in domain-specific terminology, knowledge distribution across specialties, and imbalanced data availability. We introduce MTTL-ClinicalBERT, a symmetrical multi-task transfer learning framework that harmonizes knowledge sharing across diverse medical specialties while maintaining balanced performance. Our approach addresses the fundamental problem of symmetry in knowledge transfer through three innovative components: (1) an adaptive knowledge distillation mechanism that creates symmetrical information flow between related medical domains while preventing negative transfer; (2) a bidirectional hierarchical attention architecture that establishes symmetry between local terminology analysis and global contextual understanding; and (3) a dynamic task-weighting strategy that maintains equilibrium in the learning process across asymmetrically distributed medical specialties. Extensive experiments on the MTSamples dataset demonstrate that our symmetrical approach consistently outperforms asymmetric baselines, achieving average improvements of 7.2% in accuracy and 6.8% in F1-score across five major specialties. The framework’s knowledge transfer patterns reveal a symmetric similarity matrix between specialties, with strongest bidirectional connections between cardiovascular/pulmonary and surgical domains (similarity score 0.83). Our model demonstrates remarkable stability and balance in low-resource scenarios, maintaining over 85% classification accuracy with only 30% of training data. The proposed framework not only advances clinical text classification through its symmetrical design but also provides valuable insights into balanced information sharing between different medical domains, with broader implications for symmetrical knowledge transfer in multi-domain machine learning systems. Full article

The thick cloud layer enshrouding Venus influences its thermal balance and climate evolution. However, our knowledge of total optical depth, spatial and temporal variations in the clouds is limited. We present the first complete study of the SPICAV IR spectrometer observations in the 1.28- and 1.31-µm atmospheric transparency windows during the Venus Express mission in 2006–2014. The nadir spectra were analyzed with one-dimensional multiple scattering radiative transfer model to obtain the variability of total cloud opacity on the Venus night side. The optical depth recomputed to 1 µm averages 36.7 with a standard deviation of 6.1. Cloud opacity depends on latitude, with a minimum at 50–55° N. In the Southern Hemisphere, this latitude dependence is less pronounced due to the reduced spatial resolution of the experiment, determined by the eccentricity of the spacecraft’s orbit. Cloud opacity exhibits strong variability at short time scales, mostly in the range of 25–50. The variability is more pronounced in the equatorial region. The lack of imaging capability limits the quantitative characterization of the periodicity. No persistent longitude or local time trends were detected. Full article

The Space-Air-Ground Integrated Network (SAGIN) has emerged as a core architecture for future intelligent communication due to its wide-area coverage and dynamic heterogeneous characteristics. However, its high latency, dynamic topology, and privacy–security challenges severely constrain the application of Federated Learning (FL). This paper proposes a Privacy-Preserving Federated Learning framework for SAGIN (PPFL-SAGIN), which for the first time integrates differential privacy, adaptive transfer learning, and bi-level reinforcement learning to systematically address data heterogeneity, device dynamics, and privacy leakage in SAGINs. Specifically, (1) an adaptive knowledge-sharing mechanism based on transfer learning is designed to balance device heterogeneity and data distribution divergence through dynamic weighting factors; (2) a bi-level reinforcement learning device selection strategy is proposed, combining meta-learning and hierarchical attention mechanisms to optimize global–local decision-making and enhance model convergence efficiency; (3) dynamic privacy budget allocation and robust aggregation algorithms are introduced to reduce communication overhead while ensuring privacy. Finally, experimental evaluations validate the proposed method. Results demonstrate that PPFL-SAGIN significantly outperforms baseline solutions such as FedAvg, FedAsync, and FedAsyncISL in terms of model accuracy, convergence speed, and privacy protection strength, verifying its effectiveness in addressing privacy preservation, device selection, and global aggregation in SAGINs. Full article

This study examines how generative AI (GAI) impacts primary students’ in-depth learning, focusing on critical thinking and prior knowledge. A quasi-experiment involved 163 sixth-graders divided into three groups: a control group (lecture-based instruction) and two experimental groups using GAI as a cognitive tool (materials generation) or thinking tool (critical analysis), in which 126 participants successfully completed all the tests and were included in the analysis. ANOVA revealed the thinking-tool group and cognitive-tool group both outperformed the control group in in-depth learning, which was reflected by the knowledge transfer. Hierarchical regression showed students’ critical thinking skills and use of generative artificial intelligence significantly contributed to their in-depth learning, while prior knowledge did not. Further analysis found that significant interaction effects existed between the use of generative artificial intelligence and critical thinking skills, while no significant interaction was found between the use of generative artificial intelligence and students’ prior knowledge. In sum, critical thinking amplified GAI’s impact, while prior knowledge showed no interaction. The results suggest GAI enhances deep learning when integrated with critical thinking, reducing reliance on prior knowledge. Educators should prioritize fostering critical thinking to maximize GAI’s benefits. The findings underscore the need for pedagogical designs that balance GAI’s cognitive support with metacognitive skill development. Full article

This study proposes a multi-source unsupervised domain adaptation framework for person re-identification (ReID), addressing cross-domain feature discrepancies and label scarcity in electric power field operations. Inspired by symmetry principles in feature space optimization, the framework integrates (1) a Reverse Attention-based Feature Fusion (RAFF) module aligning cross-domain features using symmetry-guided prototype interactions that enforce bidirectional style-invariant representations and (2) a Self-Correcting Pseudo-Label Loss (SCPL) dynamically adjusting confidence thresholds using entropy symmetry constraints to balance source-target domain knowledge transfer. Experiments demonstrate 92.1% rank-1 accuracy on power industry benchmarks, outperforming DDAG and MTL by 9.5%, with validation confirming robustness in operational deployments. The symmetric design principles significantly enhance model adaptability to inherent symmetry breaking caused by heterogeneous power grid environments. Full article