Search Results (415)

Modern energy storage systems for both power and transportation are highly related to lithium-ion batteries (LIBs). However, their safety depends on a potentially hazardous failure mode known as thermal runaway (TR). Predicting and classifying TR causes can widely enhance the safety of power and transportation systems. This paper presents an advanced machine learning method for forecasting and classifying the causes of TR. A generative model for synthetic data generation was used to handle class imbalance in the dataset. Hyperparameter optimization was conducted using Optuna for four classifiers: Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), tabular network (TabNet), and Extreme Gradient Boosting (XGBoost). A three-fold cross-validation approach was used to guarantee a robust evaluation. An open-source database of LIB failure events is used for model training and testing. The XGBoost model outperforms the other models across all TR categories by achieving 100% accuracy and a high recall (1.00). Model results were interpreted using SHapley Additive exPlanations analysis to investigate the most significant factors in TR predictors. The findings show that important TR indicators include energy adjusted for heat and weight loss, heater power, average cell temperature upon activation, and heater duration. These findings guide the design of safer battery systems and preventive monitoring systems for real applications. They can help experts develop more efficient battery management systems, thereby improving the performance and longevity of battery-operated devices. By enhancing the predictive knowledge of temperature-driven failure mechanisms in LIBs, the study directly advances thermal analysis and energy storage safety domains. Full article

As a crucial component of account protection system evaluation and intrusion detection, the advancement of password guessing technology encounters challenges due to its reliance on password data. In password guessing research, there is a conflict between the traditional models’ need for large training samples and the limitations on accessing password data imposed by privacy protection regulations. Consequently, security researchers often struggle with the issue of having a very limited password set from which to guess. This paper introduces a small-sample password guessing technique that enhances cross-domain features. It analyzes the password set using probabilistic context-free grammar (PCFG) to create a list of password structure probabilities and a dictionary of password fragment probabilities, which are then used to generate a password set structure vector. The method calculates the cosine similarity between the small-sample password set B from the target area and publicly leaked password sets $A_i$ using the structure vector, identifying the set $A_{max}$ with the highest similarity. This set is then utilized as a training set, where the features of the small-sample password set are enhanced by modifying the structure vectors of the training set. The enhanced training set is subsequently employed for PCFG password generation. The paper uses hit rate as the evaluation metric, and Experiment I reveals that the similarity between B and $A_i$ can be reliably measured when the size of B exceeds 150. Experiment II confirms the hypothesis that a higher similarity between $A_i$ and B leads to a greater hit rate of $A_i$ on the test set of B, with potential improvements of up to 32% compared to training with B alone. Experiment III demonstrates that after enhancing the features of $A_{max}$, the hit rate for the small-sample password set can increase by as much as 10.52% compared to previous results. This method offers a viable solution for small-sample password guessing without requiring prior knowledge. Full article

Narrative review synthesizes the most current literature on the SARS-CoV-2 XEC variant, focusing on its genomic evolution, immune evasion characteristics, epidemiological dynamics, and public health implications. To achieve this, we conducted a structured search of the literature of peer-reviewed articles, preprints, and official surveillance data from 2023 to early 2025, prioritizing virological, clinical, and immunological reports related to XEC and its parent lineages. Defined by the distinctive spike protein mutations, T22N and Q493E, XEC exhibits modest reductions in neutralization in vitro, although current evidence suggests that mRNA booster vaccines, including those targeting JN.1 and KP.2, retain cross-protective efficacy against symptomatic and severe disease. The XEC strain of SARS-CoV-2 has drawn particular attention due to its increasing prevalence in multiple regions and its potential to displace other Omicron subvariants, although direct evidence of enhanced replicative fitness is currently lacking. Preliminary analyses also indicated that glycosylation changes at the N-terminal domain enhance infectivity and immunological evasion, which is expected to underpin the increasing prevalence of XEC. The XEC variant, while still emerging, is marked by a unique recombination pattern and a set of spike protein mutations (T22N and Q493E) that collectively demonstrate increased immune evasion potential and epidemiological expansion across Europe and North America. Current evidence does not conclusively associate XEC with greater disease severity, although additional research is required to determine its clinical relevance. Key knowledge gaps include the precise role of recombination events in XEC evolution and the duration of cross-protective T-cell responses. New research priorities include genomic surveillance in undersampled regions, updated vaccine formulations against novel spike epitopes, and long-term longitudinal studies to monitor post-acute sequelae. These efforts can be augmented by computational modeling and the One Health approach, which combines human and veterinary sciences. Recent computational findings (GISAID, 2024) point to the potential of XEC for further mutations in under-surveilled reservoirs, enhancing containment challenges and risks. Addressing the potential risks associated with the XEC variant is expected to benefit from interdisciplinary coordination, particularly in regions where genomic surveillance indicates a measurable increase in prevalence. Full article

We present a new method for trend analysis that integrates traditional foresight techniques with advanced data processing and artificial intelligence. It addresses the challenge of analyzing large volumes of information while preserving expert insight. The hybrid methodology combines computational analysis with expert validation across four phases: literature review, information systematization, trend identification, and analysis. Tools like Voyant Tools 2.6.18 and NotebookLMare used for semantic and statistical exploration. Among them, we highlight the use of the Defly Compass tool, a natural language processing tool based on semantic projections and developed by our team. The method produces mixed results, including both conceptual conclusions and quantifiable, reproducible outcomes adaptable to diverse contexts. Comparative case studies in agriculture, education, and public health identified key patterns within and across sectors. Cross-domain validation revealed universal trends such as digital infrastructure, data integration, and equity. Designed for accessibility, the method enables small, non-specialized teams to combine computational tools with expert knowledge for strategic decision making in complex environments. Full article

The innovation demand and supply matching requires an accurate and time-consuming analysis of patents and the identification of their technological domains; since these tasks can be particularly challenging, this is why recent studies have evaluated the possibility of adopting Artificial Intelligence based on NLP techniques. Here, we present an automated workflow for patent analysis and classification devoted to the Italian patent scenario. High-quality data from the online platform KnowledgeShare (KS) were investigated: KS is the first patent management platform on the Italian innovation scene. A not secondary aspect consisted in determining which words mostly influenced patent classification, thus characterizing the corresponding research areas. Several models were compared to ensure the workflow’s robustness; Logistic Regression (LR) resulted in the best-performing model, and its performance compared well with the State of the Art. For each technological domain in the KS database, we evaluated and discussed its characteristic words; furthermore, a further analysis was focused on explaining why some domains, such as “Packaging” and “Environment,” were particularly confounding. This last aspect is of paramount importance to identify cross-contamination effects among research areas. Full article

Food security constitutes a fundamental pillar of future sustainable development. A systematic evaluation of the impact of urban expansion on food security is critical to advancing the United Nations Sustainable Development Goals (SDGs), particularly “Zero Hunger” (SDG 2). Drawing on bibliographic data from the Web of Science Core Collection, this study employs the bibliometrix package in R to conduct a comprehensive bibliometric analysis of the literature on the “urban expansion–food security” nexus spanning from 1982 to 2024. The analysis focuses on knowledge production, collaborative structures, and thematic research trends. The results indicate the following: (1) The publication trajectory in this field exhibits a generally increasing trend with three distinct phases: an incubation period (1982–2000), a development phase (2001–2014), and a phase of rapid growth (2015–2024). Land Use Policy stands out as the most influential journal in the domain, with an average citation rate of 43.5 per article. (2) China and the United States are the leading contributors in terms of publication output, with 3491 and 1359 articles, respectively. However, their international collaboration rates remain relatively modest (0.19 and 0.35) and considerably lower than those observed for the United Kingdom (0.84) and Germany (0.76), suggesting significant potential for enhanced global research cooperation. (3) The major research hotspots cluster around four core areas: urban expansion and land use dynamics, agricultural systems and food security, environmental and climate change, and socio-economic and policy drivers. These focal areas reflect a high degree of interdisciplinary integration, particularly involving land system science, agroecology, and socio-economic studies. Collectively, the field has established a relatively robust academic network and coherent knowledge framework. Nonetheless, it still confronts several limitations, including geographical imbalances, fragmented research scales, and methodological heterogeneity. Future efforts should emphasize cross-regional, interdisciplinary, and multi-scalar integration to strengthen the systematic understanding of urban expansion–food security interactions, thereby informing global strategies for sustainable development. Full article

Remote sensing visual grounding, a pivotal technology bridging natural language and high-resolution remote sensing images, holds significant application value in disaster monitoring, urban planning, and related fields. However, it faces critical challenges due to the inherent scale heterogeneity, semantic complexity, and annotation scarcity of remote sensing data. This paper first reviews the development history of remote sensing visual grounding, providing an overview of the basic background knowledge, including fundamental concepts, datasets, and evaluation metrics. Then, it categorizes methods by whether they employ large language models as a pedestal, and provides in-depth analyses of the innovations and limitations of Transformer-based and multimodal large language model-based methods. Furthermore, focusing on remote sensing image characteristics, it discusses cutting-edge techniques such as cross-modal feature fusion, language-guided visual optimization, multi-scale, and hierarchical feature processing, open-set expansion and efficient fine-tuning. Finally, it outlines current bottlenecks and proposes valuable directions for future research. As the first comprehensive review dedicated to remote sensing visual grounding, this work is a reference resource for researchers to grasp domain-specific concepts and track the latest developments. 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

Underwater acoustic target recognition (UATR) is challenged by complex marine noise, scarce labeled data, and inadequate multi-scale feature extraction in conventional methods. This study proposes DART-MT, a semi-supervised framework that integrates a Dual Attention Parallel Residual Network Transformer with a mean teacher paradigm, enhanced by domain-specific prior knowledge. The architecture employs a Convolutional Block Attention Module (CBAM) for localized feature refinement, a lightweight New Transformer Encoder for global context modeling, and a novel TriFusion Block to synergize spectral–temporal–spatial features through parallel multi-branch fusion, addressing the limitations of single-modality extraction. Leveraging the mean teacher framework, DART-MT optimizes consistency regularization to exploit unlabeled data, effectively mitigating class imbalance and annotation scarcity. Evaluations on the DeepShip and ShipsEar datasets demonstrate state-of-the-art accuracy: with 10% labeled data, DART-MT achieves 96.20% (DeepShip) and 94.86% (ShipsEar), surpassing baseline models by 7.2–9.8% in low-data regimes, while reaching 98.80% (DeepShip) and 98.85% (ShipsEar) with 90% labeled data. Under varying noise conditions (−20 dB to 20 dB), the model maintained a robust performance (F1-score: 92.4–97.1%) with 40% lower variance than its competitors, and ablation studies validated each module’s contribution (TriFusion Block alone improved accuracy by 6.9%). This research advances UATR by (1) resolving multi-scale feature fusion bottlenecks, (2) demonstrating the efficacy of semi-supervised learning in marine acoustics, and (3) providing an open-source implementation for reproducibility. In future work, we will extend cross-domain adaptation to diverse oceanic environments. Full article

As industries become increasingly dependent on cyber-physical systems (CPSs), failures within these systems can cause significant operational disruptions, underscoring the critical need for effective Prognostics and Health Management (PHM). The large volume of data generated by CPSs has made deep learning (DL) methods an attractive solution; however, imbalanced datasets and the limited availability of fault-labeled data continue to hinder their effective deployment in real-world applications. To address these challenges, this paper proposes a transfer learning approach using a pre-trained transformer architecture to enhance fault detection performance in CPSs. A streamlined transformer model is first pre-trained on a large-scale source dataset and then fine-tuned end-to-end on a smaller dataset with a differing data distribution. This approach enables the transfer of diagnostic knowledge from controlled laboratory environments to real-world operational settings, effectively addressing the domain shift challenge commonly encountered in industrial CPSs. To evaluate the effectiveness of the proposed method, extensive experiments are conducted on publicly available datasets generated from a laboratory-scale replica of a modern industrial water purification facility. The results show that the model achieves an average F1-score of 93.38% under K-fold cross-validation, outperforming baseline models such as CNN and LSTM architectures, and demonstrating the practicality of applying transformer-based transfer learning in industrial settings with limited fault data. To enhance transparency and better understand the model’s decision process, SHAP is applied for explainable AI (XAI). Full article

Artificial intelligence and bioacoustics represent a paradigm shift in non-invasive poultry welfare monitoring through advanced vocalization analysis. This comprehensive systematic review critically examines the transformative evolution from traditional acoustic feature extraction—including Mel-Frequency Cepstral Coefficients (MFCCs), spectral entropy, and spectrograms—to cutting-edge deep learning architectures encompassing Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, attention mechanisms, and groundbreaking self-supervised models such as wav2vec2 and Whisper. The investigation reveals compelling evidence for edge computing deployment via TinyML frameworks, addressing critical scalability challenges in commercial poultry environments characterized by acoustic complexity and computational constraints. Advanced applications spanning emotion recognition, disease detection, and behavioral phenotyping demonstrate unprecedented potential for real-time welfare assessment. Through rigorous bibliometric co-occurrence mapping and thematic clustering analysis, this review exposes persistent methodological bottlenecks: dataset standardization deficits, evaluation protocol inconsistencies, and algorithmic interpretability limitations. Critical knowledge gaps emerge in cross-species domain generalization and contextual acoustic adaptation, demanding urgent research prioritization. The findings underscore explainable AI integration as essential for establishing stakeholder trust and regulatory compliance in automated welfare monitoring systems. This synthesis positions acoustic AI as a cornerstone technology enabling ethical, transparent, and scientifically robust precision livestock farming, bridging computational innovation with biological relevance for sustainable poultry production systems. Future research directions emphasize multi-modal sensor integration, standardized evaluation frameworks, and domain-adaptive models capable of generalizing across diverse poultry breeds, housing conditions, and environmental contexts while maintaining interpretability for practical farm deployment. Full article

E-commerce live streaming in China has become a major retail channel, yet hosts often employ subtle phonetic or semantic “morphs” to evade moderation and make unsubstantiated claims, posing risks to consumers. To address this, we study the Live Auditory Morph Resolution (LiveAMR) task, which restores morphed speech transcriptions to their true forms. Building on prior text-based morph resolution, we propose an LLM-enhanced training framework that mines three types of explanation knowledge—predefined morph-type labels, LLM-generated reference corrections, and natural-language rationales constrained for clarity and comprehensiveness—from a frozen large language model. These annotations are concatenated with the original morphed sentence and used to fine-tune a lightweight T5 model under a standard cross-entropy objective. In experiments on two test sets (in-domain and out-of-domain), our method achieves substantial gains over baselines, improving $F_{0.5}$ by up to 7 pp in-domain (to 0.943) and 5 pp out-of-domain (to 0.799) compared to a strong T5 baseline. These results demonstrate that structured LLM-derived signals can be mined without fine-tuning the LLM itself and injected into small models to yield efficient, accurate morph resolution. Full article

Although cooperative perception enhances situational awareness by enabling vehicles to share intermediate features, real-world deployment faces challenges due to heterogeneity in sensor modalities, architectures, and encoder parameters across agents. These domain gaps often result in semantic inconsistencies among the shared features, thereby degrading the quality of feature fusion. Existing approaches either necessitate the retraining of private models or fail to adapt to newly introduced agents. To address these limitations, we propose PnPDA⁺, a unified and modular domain adaptation framework designed for heterogeneous multi-vehicle cooperative perception. PnPDA⁺ consists of two key components: a Meta Feature Extraction Network (MFEN) and a Plug-and-Play Domain Adapter (PnPDA). MFEN extracts domain-aware and frame-aware meta features from received heterogeneous features, encoding domain-specific knowledge and spatial-temporal cues to serve as high-level semantic priors. Guided by these meta features, the PnPDA module performs adaptive semantic conversion to enhance cross-agent feature alignment without modifying existing perception models. This design ensures the scalable integration of emerging vehicles with minimal fine-tuning, significantly improving both semantic consistency and generalization. Experiments on OPV2V show that PnPDA⁺ outperforms state-of-the-art methods by 4.08% in perception accuracy while preserving model integrity and scalability. Full article

Large-language-model (LLM) APIs demonstrate impressive reasoning capabilities, but their size, cost, and closed weights limit the deployment of knowledge-aware AI within biomedical research groups. At the other extreme, standard attention-based neural language models (SANLMs)—including encoder–decoder architectures such as Transformers, Gated Recurrent Units (GRUs), and Long Short-Term Memory (LSTM) networks—are computationally inexpensive. However, their capacity for semantic reasoning in noisy, open-vocabulary knowledge bases (KBs) remains unquantified. Therefore, we investigate whether compact SANLMs can (i) reason over hybrid OpenIE-derived KBs that integrate commonsense, general-purpose, and non-communicable-disease (NCD) literature; (ii) operate effectively on commodity GPUs; and (iii) exhibit semantic coherence as assessed through manual linguistic inspection. To this end, we constructed four training KBs by integrating ConceptNet (600k triples), a 39k-triple general-purpose OpenIE set, and an 18.6k-triple OpenNCDKB extracted from 1200 PubMed abstracts. Encoder–decoder GRU, LSTM, and Transformer models (1–2 blocks) were trained to predict the object phrase given the subject + predicate. Beyond token-level cross-entropy, we introduced the Meaning-based Selectional-Preference Test (MSPT): for each withheld triple, we masked the object, generated a candidate, and measured its surplus cosine similarity over a random baseline using word embeddings, with significance assessed via a one-sided t-test. Hyperparameter sensitivity (311 GRU/168 LSTM runs) was analyzed, and qualitative frame–role diagnostics completed the evaluation. Our results showed that all SANLMs learned effectively from the point of view of the cross entropy loss. In addition, our MSPT provided meaningful semantic insights: for the GRUs (256-dim, 2048-unit, 1-layer): mean similarity $\left({\mu }_{sts}\right)$ of 0.641 to the ground truth vs. 0.542 to the random baseline (gap 12.1%; $p<{10}^{-180}$). For the 1-block Transformer: ${\mu }_{sts}=0.551$ vs. $0.511$ (gap 4%; $p<{10}^{-25}$). While Transformers minimized loss and accuracy variance, GRUs captured finer selectional preferences. Both architectures trained within <24 GB GPU VRAM and produced linguistically acceptable, albeit over-generalized, biomedical assertions. Due to their observed performance, LSTM results were designated as baseline models for comparison. Therefore, properly tuned SANLMs can achieve statistically robust semantic reasoning over noisy, domain-specific KBs without reliance on massive LLMs. Their interpretability, minimal hardware footprint, and open weights promote equitable AI research, opening new avenues for automated NCD knowledge synthesis, surveillance, and decision support. Full article

Background: Competency in providing high-quality, comprehensive patient care is essential for newly qualified dentists. Dental curricula are designed to equip graduates with necessary skills to develop competencies. Outplacement training has been incorporated into most dental curricula to provide broader clinical experience in a real-world situation. Methods: This cross-sectional study aimed to assess (1) the perceived confidence of final-year dental students (FYDSs) at an Australian university with reference to the Australian Dental Council (ADC) professional competencies for newly qualified dentists; (2) the association between perceived confidence and the timing of outplacement training; and (3) students’ perceptions on outplacement training in developing their competencies. Two online surveys were administered to a cohort of FYDSs at the end of the first and second trimesters. ‘Survey 1’ was based on the ADC competency requirements (2022) and assessed aims 1 and 2. The five domains assessed were (1) communication and leadership; (2) critical thinking; (3) health promotion; (4) scientific and clinical knowledge; and (5) person-centred care. ‘Survey 2’ assessed students’ perception on outplacement training and was administered towards the completion of their outplacement to assess aim 3. Results: Students’ perceived competency levels were high across all domains. Those with prior tertiary education were more confident in communication and leadership, health promotion, and scientific and clinical knowledge than students with secondary qualifications. The perceived confidence in professional competencies among FYDSs had no significant association (p > 0.05) with the location or the sequence of outplacement. The thematic analysis of survey 2 responses reported the guidance and constructive feedback received from supervisors while managing cases in a real-life setup as a significant contributor to their confidence development. Conclusions: FYDSs reported a satisfactory level of perceived confidence in professional competencies. Both in-house training and outplacement equally improve the levels of competency development. FYDSs perceive outplacement training as a positive integral component in the development of skills. Full article