929 papers
This paper presents an AI-driven profiling model within a cyber range environment that continuously evaluates learner performance to generate automated, personalized feedback and adaptive learning paths, thereby addressing key challenges in scalability and personalization for cybersecurity education.
This paper proposes a lightweight, real-time semantic front-end that integrates periodic instance segmentation with a registration-driven Gaussian mapping pipeline to effectively remove dynamic objects and enhance the robustness of RGB-D SLAM in dynamic environments.
This study presents LV-YOLO, a two-stage deep learning framework combining MobileNetV3 for global screening and an optimized YOLOv11 model with novel attention and upsampling modules, which effectively and accurately detects lymphovascular invasion in gastric cancer whole slide images to assist pathologists in clinical risk stratification.
This paper demonstrates that an unsupervised framework leveraging frozen 2D vision foundation model features, when applied to canonical depth renderings of LiDAR tree crowns, organizes individual tree species more coherently than both hand-crafted geometric descriptors and frozen 3D foundation models, while rigorously controlling for acquisition-source confounding to validate the superiority of pre-trained visual representations over native 3D geometry in this domain.
SpectralMLP introduces a novel hypergraph neural network framework that replaces traditional aggregation with an efficient, message-passing-free dual-frequency spectral filtering mechanism using learnable Jacobi polynomial bases, achieving state-of-the-art performance across multiple benchmarks while maintaining a parameter count lower than most existing methods.
This study presents a controlled benchmark comparing raw sequential, image-based (GAF), hybrid, and statistical models for anomaly detection in household electricity consumption, finding that direct temporal modeling (Raw 1D-CNN) outperformed other approaches under synthetic perturbation protocols while emphasizing the need for cautious interpretation due to the dataset's single-household scope and lack of real-world fault annotations.
This paper presents an intelligent warehouse execution framework that integrates SAP EWM, Large Language Models, SAP BTP, and autonomous mobile robots to enable natural-language-driven task execution, achieving a 98% success rate and an average execution time of 4.7 seconds through adaptive human-robot collaboration.
This systematic literature review analyzes 97 studies to identify six key governance mechanisms for responsible AI in business, revealing a consensus on ethical principles like accountability and fairness while highlighting significant implementation challenges and research gaps that hinder effective adoption.
The paper presents ARIES-Mission, a modular vision-language system that enhances UAV mission planning by selecting the shortest feasible route from a portfolio of four candidates (raw recognition order, ant colony optimization, differential evolution, and particle swarm optimization), achieving a 24.21% reduction in total distance and near-optimal performance with a mean gap of 0.042% to the exact optimum.
This paper presents an AI-assisted framework that generates immersive VR and MR training tutorials from expert demonstrations by capturing multimodal data to create structured, step-by-step instructions with 3D avatar replays, thereby enabling scalable, instructorless industrial training.