ABSTRACT Wildfire spread prediction is a critical task in remote sensing image analysis, where accurately identifying newly expanded wildfire regions is essential for real-time monitoring and emergency response. In this study, we propose a semantic segmentation network, termed the Residual Contextual Dual Attention Network (RCDA-Net), for predicting incremental wildfire spread regions from multi-source remote sensing data. RCDA-Net integrates two attention modules, Contextual Anchor Attention (CAA) and Adaptive Graph Channel Attention (AGCA), to enhance spatial structure modelling and inter-channel dependency learning across spectral, topographic, and meteorological inputs. We construct and release a large-scale wildfire dataset covering North America, which serves as a benchmark for incremental fire spread prediction. Experimental results on this dataset show that RCDA-Net achieves an F1-score of 0.471 and an IoU of 0.308, outperforming established models such as U-Net, AttU-Net, WPN, and FU-NetCast. With 8.9 million parameters and an inference speed of 131 frames per second (fps), RCDA-Net provides a favourable balance between segmentation accuracy and computational efficiency. Ablation studies validate the complementary effects of CAA and AGCA, while additional analyses demonstrate that Dice Loss effectively mitigates class imbalance and boundary ambiguity. Robustness evaluations further indicate that wildfire masks and meteorological factors play a dominant role in predictive performance under partial input degradation. The source code is publicly available at: https://github.com/hxxAlways/RCDA-Net.

ABSTRACT Traditionally associated with physical wellness, Physical Education (PE) is now recognized as a platform for holistic education aligned with sustainable development goals (SDGs). Spanish universities are committed to integrating these goals into their curricula. In this sense, a key challenge for Education for Sustainable Development (ESD) lies in transforming teaching approaches to integrate sustainability within specific subjects like PE, where pre-service teachers’ perceptions play a vital role in shaping innovative and effective strategies for this integration. This study explored the perspectives of key informants on the integration of sustainability into the core elements of the Spanish PE curriculum. The study sample included 75 final-year Physical Activity and Sport Sciences students. Adopting a phenomenological approach, data was collected through semi-structured interviews focused on practical applications of sustainability competencies in PE. The qualitative data was analysed in three stages using Atlas.ti v.7.5.18 software: data organization, category coding, and semantic network structuring. The findings, categorized into six meta-categories – objectives, content, methods, student roles, learning environment, and assessment – highlighted the importance of cultivating awareness about the long-term consequences of actions, promoting healthy lifestyles and reducing waste. Content focused on traditional sports, values, and non-polluting transport, while teaching methods included cooperative learning, problem-solving, and roles simulation. Formative assessment, particularly self – and peer-assessment, emerged as crucial for promoting reflection and growth in sustainability competencies. In conclusion, a sustainability-focused pedagogical approach in PE requires diverse, student-centred, and reflective teaching methods to develop environmental, economic, and social competencies. From the pre-service teachers’ perspective, this study offers a foundation for creating a pedagogical model to incorporate competencies for sustainable development in PE. Future research could refine, implement, and evaluate the effectiveness of this model in fostering sustainability competences.

During 2024-2025, global emergencies triggered intense online discourse, presenting a unique opportunity to examine how cultural factors shape emotional expression and knowledge dissemination. Understanding these dynamic mechanisms is crucial for enhancing the effectiveness of digital health communication and optimizing crisis response strategies. We analyzed how cultural and linguistic contexts influence emotional expression and thematic framing in social media comments during major emergencies in 2024-2025. We uncovered cross-cultural differences in collective emotions and narrative focuses, explaining how affective stance and discourse framing jointly shape the public construction of crisis meaning. We used a cross-sectional, convergent mixed methods design. Data were collected retrospectively from X (formerly Twitter; X Corp) and Weibo (Sina Weibo) between January 1 and December 31, 2024. Using purposive sampling, we selected 5-6 representative emergency events per month based on online visibility (capped at 600 comments/event). The dataset included 19,813 comments from X and 6536 comments from Weibo. Emotions were identified using a Cross-lingual Language Model-Robustly optimized Bidirectional Encoder Representations from Transformers approach, and thematic patterns were extracted with Bidirectional Encoder Representations from Transformers Topic. Integrated Gradients was used to interpret model outputs, while clustering and network analysis were applied to visualize cross-cultural patterns. Hofstede's cultural dimensions theory helped interpret cultural influences on discourse. This mixed computational approach enabled a detailed comparison of emotional structures and thematic discourse across linguistic communities. Significant cross-platform differences were observed in emotional distribution (χ²8=8025.60; P<.001). Compared to X users, Weibo users, representing a collectivist culture, expressed concentrated negative emotions (20.37%; odds ratio [OR] 15.76, 95% CI 13.90-17.85), surprise (19.70%; OR 2.53, 95% CI 2.32-2.73), and fear (16.68%; OR 1.72, 95% CI 1.58-1.86), reflecting group-oriented anxiety and emotional contagion. In contrast, X (formerly Twitter) users in individualist contexts displayed dispersed sarcasm (43.49%; OR 55.19, 95% CI 43.95-69.21) and worry (15.30%; OR 55.27, 95% CI 34.74-87.88), indicating personalized and critical emotional styles. Topic modeling revealed dense clusters around "safety," "pray," and "resettlement" on Weibo, whereas X (formerly Twitter) comments emphasized decentralized themes of critique and responsibility. Semantic network analysis revealed a cohesive fear-prayer-rescue chain on Weibo and fragmented, debate-oriented interactions on X (formerly Twitter). Emergency discourse is not neutral but is systematically structured by cultural values that shape emotions and themes. Integrating multilingual computational and qualitative methods, we offer a replicable framework using large-scale data, moving crisis and infodemiological research beyond single-platform or survey-based approaches. Our findings advance theory-informed understanding of how cultural meaning systems translate into observable digital discourse under conditions of risk and uncertainty. They also offer practical implications for governments, public health agencies, international organizations, and digital platforms by informing culturally adaptive, platform-specific risk communication, community moderation, and crisis engagement strategies that can strengthen public trust, improve compliance with protective behaviors, and mitigate infodemic-related harms.

To achieve low-cost and flexible wheel angles measurement, we propose a novel strategy that integrates wheel segmentation network with 3D vision. In this framework, a semantic segmentation network is first employed to extract the wheel rim, followed by angle estimation through ICP-based point cloud registration. Since wheel rim extraction is closely tied to angle computation accuracy, we introduce APCS-SwinUnet, a segmentation network built on the SwinUnet architecture and enhanced with ASPP, CBAM, and a hybrid loss function. Compared with traditional image processing methods in wheel alignment, APCS-SwinUnet delivers more accurate and refined segmentation, especially at wheel boundaries. Moreover, it demonstrates strong adaptability across diverse tire types and lighting conditions. Based on the segmented mask, the wheel rim point cloud is extracted, and an iterative closest point algorithm is then employed to register the target point cloud with a reference one. Taking the zero-angle condition as the reference, the rotation and translation matrices are obtained through point cloud registration. These matrices are subsequently converted into toe and camber angles via matrix-to-angle transformation. Experimental results verify that the proposed solution enables accurate angle measurement in a cost-effective, simple, and flexible manner. Furthermore, repeated experiments further validate its robustness and stability.

Building extraction underpins land-use assessment, urban planning, and disaster mitigation, yet dense urban scenes still cause missed small objects, target adhesion, and ragged contours. We present High-Resolution-Mamba (HR-Mamba), a high-resolution semantic segmentation network that augments a High-Resolution Network (HRNet) parallel backbone with edge-aware and sequence-state modeling. A Canny-enhanced, median-filtered stem stabilizes boundaries under noise; Involution-based residual blocks capture position-specific local geometry; and a Mamba-based State Space Models (Mamba-SSM) global branch captures cross-scale long-range dependencies with linear complexity. Training uses a composite loss of binary cross entropy (BCE), Dice loss, and Boundary loss, with weights selected by joint grid search. We further design a feature-driven adaptive post-processing pipeline that includes geometric feature analysis, multi-strategy simplification, multi-directional regularization, and topological consistency verification to produce regular, smooth, engineering-ready building outlines. On dense urban imagery, HR-Mamba improves F1-score from 80.95% to 83.93%, an absolute increase of 2.98% relative to HRNet. We conclude that HR-Mamba jointly enhances detail fidelity and global consistency and offers a generalizable route for high-resolution building extraction in remote sensing.

Creativity, the fountain of youth: Association between creativity and semantic memory networks across the lifespan.

The role of semantic features in word production.

The impact of distractor processing on semantic memory retrieval: The role of interference-by-process and inhibition.

Object dynamic recognition and grasping location via lightweight semantic attention network with learnable boundary vectors

ELITE-Seg: An explainable and lightweight semantic terrain segmentation network based on SCSE-gated EfficientNet

Joint Semantic Information Extraction and Resource Allocation in User-Centric Semantic Communication Networks

Recent years have witnessed the remarkable progress of 3D multi-modality object detection methods based on the Bird's-Eye-View (BEV) perspective. However, most of them overlook the complementary interaction and guidance between LiDAR and camera. In this work, we propose a novel multi-modality 3D objection detection method, with multi-guided global interaction and LiDAR-guided adaptive fusion, named MGAF. Specifically, we introduce sparse depth guidance (SDG) and LiDAR occupancy guidance (LOG) to generate 3D features with sufficient depth and spatial information. The designed semantic segmentation network captures category and orientation prior information for raw point clouds. In the following, an Adaptive Fusion Dual Transformer (AFDT) is developed to adaptively enhance the interaction of different modal BEV features from both global and bidirectional perspectives. Meanwhile, additional downsampling with sparse height compression and multi-scale dual-path transformer (MSDPT) are designed in order to enlarge the receptive fields of different modal features. Finally, a temporal fusion module is introduced to aggregate features from previous frames. Notably, the proposed AFDT is general, which also shows superior performance on other models. Our framework has undergone extensive experimentation on the large-scale nuScenes dataset, Waymo Open Dataset, and long-range Argoverse2 dataset, consistently demonstrating state-of-the-art performance.

Does nonsense make sense? A springboard to studying dynamic reconfiguration of large-scale networks during semantic and intonation speech processing.

InSegNet: Semantic segmentation-based lane detection network for clear and occluded roads

Understanding semantic impairments in schizophrenia from a predictive coding perspective.

An optimal solution for reactive power re-dispatch agency using open-source restructuring learning semantic networks

The purpose of this study is to empirically examine how teacher agency is represented through linguistic patterns and semantic structures within educational discourses. To this end, a corpus of 2,335 English-language texts—including academic articles, national education policy documents, and reports from international organizations published between 2000 and 2025—was compiled and analyzed using corpus-based semantic techniques to investigate word co-occurrence relationships and community structures. The analysis identified four major thematic clusters shaping discourses on teacher agency. First, the “Teacher Professionalism and Identity” cluster highlights teachers’ professional growth and the development of autonomous identities through reflective practice and experiential learning. Second, the “Educational Policy and Systems” cluster illustrates teachers’ transformative roles in reconstructing institutional and curricular frameworks as agents of pedagogical change. Third, the “Teachers’ Experience and Perception” cluster demonstrates that teachers’ emotions, experiences, and perceptions form the psychological foundations of agency. Fourth, the “Teacher Leadership and Support” cluster shows that teacher agency is reinforced through relational leadership grounded in trust, collaboration, and collective efficacy. Taken together, these findings indicate that teacher agency is an ecological construct shaped by the interplay of personal, structural, relational, and affective dimensions. This study contributes to the field by systematically organizing the multifaceted concept of teacher agency and by visualizing its semantic networks across academic, policy, and international discourses through a corpus-driven analytical approach.

Abstract. Accurate, topologically consistent building footprints are essential for building reconstruction and GIS applications. But high-resolution orthophotos often contain occlusions (trees, cast shadows, etc.) or dense roof structures that challenge pixel-based segmentation and polygonization. In recent years, Line Segment Detection (LSD) networks have gained popularity as they can directly extract vectorized building footprints. This study benchmarks three line-segment detection (LSD) networks - L-CNN, ULSD, and F-Clip - against a strong semantic segmentation network - DeepLabV3+ - for building footprint extraction from very high resolution orthophotos across multiple regions with varied built-up morphology. Our evaluation on the considered urban areas revealed that LSD approaches generally deliver cleaner boundaries and more reliable roof topology than segmentation methods, whose high pixel scores mask boundary breaks. These findings indicate that when polygonal fidelity and downstream GIS usability are priorities, LSD pipelines could be superior for vectorized building footprint extraction compared to segmentation methods.

Purpose This study aims to examine how the Myeongryang Naval Battle Festival reconstructs and reproduces ethnic identity amid the structural crises of local extinction and the weakening of regional identity. The core research question is to identify how symbolic elements of war, victory, heroism, and resistance embedded in the festival function as discourses shaping the ethnic identity of the local community. Methods Using Text-mining techniques were employed to extract from media coverage and SNS’s 2,680 texts, key terms, conduct frequency and TF–IDF analyses, and semantic network analysis was used to examine relational structures among terms. LDA topic modeling was subsequently applied to derive three thematic discourse structures related to the festival. Results Core terms, such as Myeongryang Naval Battle, festival, Yi Sun-sin etc. formed the central axis, indicating that the festival narrative is structured around place, historical figures, and events. Degree centrality showed that regional names function not merely as background elements but as binding nodes connecting policy, events, and tourism, demonstrating that the festival serves as a mechanism for reproducing both regional and ethnic identities. Topic modeling revealed three discourse structures: ethnic-identity discourse, tourism-experience discourse and regional development. Conclusion. The Myeongryang Naval Battle Festival reinforces ethnic identity by re-enacting heroic narratives and victory memories, while simultaneously contributing to regional economic revitalization through tourism-oriented experiential content. Furthermore, its integration with modern tourism infrastructure positions the festival as a key platform for regional development and a strategic response to local extinction.

To support campus security governance, especially campus traffic safety management, many ternary extraction techniques in knowledge graphs rely on character-level text analysis; however, the differences between word semantics and overall word meanings often lead to ambiguity and overlap in ternary extraction. Traditional methods are difficult to effectively manage overlapping relationships in text, which seriously damages the flexibility and extraction accuracy of the dataset. In addition, entity separation significantly affects entity relationship triplet extraction, and weak associations between remote entities often blur entity boundaries and reduce recall rates. This study offers the MGRel entity relationship extraction model, which integrates dependent syntactic analysis with a graph neural network to address the issues above and enhance knowledge acquisition for campus security scenarios. Firstly, by incorporating the dependent syntactic parser alongside the dual analysis mechanism of global semantic dependency and syntactic dependency, it effectively captures long-distance semantic associations and enhances entity relationship recognition accuracy; secondly, it devises the architecture of a hierarchical semantic graph convolutional neural network to facilitate the fine-grained extraction of deep implied semantic features among entities; finally, the attention-driven multi-feature fusion module is presented to improve the discriminative capacity of the ternary classifier via a noise filtering approach. The experimental results on three core general-purpose benchmark datasets-NYT, WebNLG, and DuIE-show that the F1 score of this model increases by 1.3, 0.4 and 3.2%, respectively, compared with the current optimal model, demonstrating a considerable advantage over the comparative techniques and potential value for campus security-oriented campus traffic safety applications.