898 papers
This paper proposes DuTLR-Net, a deep unrolled tensor low-rank network that integrates physical priors via a Cross-Scale Manifold Alignment module and an ADMM-based iterative optimization backbone to achieve superior, interpretable hyperspectral image classification with high spatial-spectral consistency, even in small-sample scenarios.
This paper proposes a sustainable blockchain framework for carbon markets that combines a hybrid Proof-of-Stake and Proof-of-Work consensus mechanism for small-to-medium applications with the Osmosis DeFi platform for large-scale operations, demonstrating significant improvements in energy efficiency, throughput, and latency compared to traditional systems.
This paper proposes a source-verification approach for detecting email-based phishing that combines NLP to extract target institutions with web mining to compare provided URLs and content against official websites, achieving 96% accuracy and outperforming existing detection services.
This study presents an integrated Augmented Reality and embedded electronics toolkit that significantly enhances undergraduate STEM students' understanding of complex network graphs, particularly benefiting those with low initial confidence, by combining synchronized physical-virtual visualizations with tactile interaction and contextual overlays.
This paper proposes NSP-ML, a multimodal learning framework that integrates visual data with normality-guided and spatiotemporal prior-aware textual logs to significantly improve the recognition of context-dependent abnormal behaviors in campus environments, achieving state-of-the-art performance on the CABRH8 dataset.
This paper proposes IoT-TrustChain, a resource-efficient blockchain-based access control framework that integrates dynamic delegation and event-aware policies to significantly reduce decision latency and gas consumption while enhancing real-time responsiveness for constrained IoT environments.
This study demonstrates that image-embedded prompt injection poses a significant cross-vendor security risk to dental vision-language models, revealing that while all tested commercial and open-source systems are vulnerable, OCR-based sanitization and provenance-aware governance strategies can effectively mitigate these attacks while preserving clinical diagnostic accuracy.
This paper proposes a novel boundary-aware deep affine attention network that integrates boundary word-pair relation modeling, dynamic convolution, and deep affine attention to effectively address challenges in entity boundary identification and long-range dependency modeling, achieving state-of-the-art performance on both flat and nested named entity recognition tasks across five benchmark datasets.
This paper introduces and evaluates Bernardes Branch-Aware Pythagorean Residual Fields as a specialized, non-manufacturing geometry-processing diagnostic tool designed to generate branch-compatible contour cues for exploratory mesh-envelope analysis, while explicitly acknowledging its limitations regarding coordinate dependence, resolution sensitivity, and lack of fabrication guarantees compared to standard distance fields.
This study proposes a deep learning framework using YOLO architectures to detect and classify eight driver activities, demonstrating that the YOLOv8n model achieves superior real-time performance with 96.3% mAP on a dataset of 5,422 images, thereby offering an effective solution for enhancing road safety in intelligent transportation systems.