2579 papers
This paper proposes CAFM, a cross-modal local alignment fusion method that utilizes local window attention, bottleneck compression, and symmetric contrastive learning to effectively integrate RGB and 3D point cloud features for superior industrial anomaly detection and localization, achieving state-of-the-art performance on the MVTec 3D-AD dataset.
This paper introduces a computationally efficient minimal solver for the P1P problem with affine correspondences (P1AC) that decomposes the task into a canonicalization step and a single quadratic equation, leveraging the equivalence between affine correspondences and gradient fields to enable applications with dense, isometry-invariant descriptor maps while providing a comprehensive analysis of degenerate and failure cases.
Grounded in the reward-learning framework, this study develops and evaluates a theory-informed generative AI agent that successfully fosters both state and trait science curiosity among university students by providing cognitive, affective, and protective support that closes curiosity loops more effectively than self-questioning alone.
This paper introduces ThyroGuard-Net, a hybrid EfficientNet–Swin Transformer framework trained on the multi-center TN3K dataset with sensitivity-weighted loss, evidential uncertainty, and TI-RADS-based explainability to achieve robust, generalizable, and clinically calibrated thyroid nodule classification with improved sensitivity over existing methods.
Graph2Path is a hierarchical, reactivity-guided template policy that enhances multi-step retrosynthesis planning by integrating atom- and bond-activity supervision into graph-encoded states, achieving superior exact-route accuracy and success rates compared to existing methods like RetroSynFormer and AiZynthFinder, albeit with increased inference time.
This paper proposes ESW-YOLO, a lightweight small-object detection framework based on YOLO11n that integrates Dynamic Snake Convolution, an enhanced BiFPN with an additional high-resolution head and iEMA attention, and WIoU loss to significantly improve detection accuracy on small and elongated targets while maintaining a comparable parameter count.
This paper proposes a novel, interpretable hybrid deep learning framework that integrates PSO, GA, and Bayesian Optimization to tune an Artificial Neural Network, achieving 99.60% accuracy in predicting student placement outcomes while utilizing SHAP and LIME to provide transparent, actionable insights for educational and industry alignment.
This paper proposes DT-STGNN-FedRL, a digital-twin-coordinated framework that integrates spatio-temporal graph learning for predictive resource forecasting with federated reinforcement learning to achieve proactive, distributed multi-resource management in 6G cellular-edge networks, significantly reducing latency and SLA violations while improving throughput and efficiency.
This paper exposes how machine learning models for predicting athletes' maximal lifts suffer from significant accuracy inflation due to concurrent-lift data leakage, and proposes a closed-form diagnostic to quantify this leakage severity alongside a cost-effective generative model that provides calibrated, physiologically consistent predictions without relying on sibling lift data.
This paper proposes an explainable hybrid deep learning framework that fuses features from six pre-trained CNN models to achieve 97.87% accuracy in classifying seven brain tumor categories, validated through visual explanation techniques and external testing.