2786 papers
This paper proposes PT-LGBM, an end-to-end interpretable cross-domain model that integrates geometry-partition sampling, feature alignment, and an optimized LightGBM classifier to achieve high-accuracy, reliable fault diagnosis for high-speed train bearings under conditions of class imbalance and domain shift.
This survey outlines a comprehensive standards-to-deployment framework for transitioning from classical to post-quantum cryptography, addressing design constraints, proposing an iterative migration pipeline, and detailing integration strategies across key protocols like TLS, IPsec, and IoT while highlighting ongoing interoperability and operational challenges.
This study proposes the Colonial Competitive Grey Wolf Optimizer (CCGWO), a hybrid algorithm that integrates socio-political evolution principles to overcome local optima limitations and effectively solve constrained economic load dispatch problems across various power systems.
This study demonstrates that training sentiment classifiers on a larger, majority-vote dataset of Indonesian coastal tourism reviews yields higher performance than a strictly unanimous-consensus dataset, though the results suggest that increased data volume and class balance, rather than the retention of ambiguous instances alone, may be the primary drivers of improvement while IndoBERT effectively reduces systematic misclassification of neutral reviews.
This paper introduces the first supervised four-modality benchmark and session-disjoint evaluation protocol for robotic welding defect classification, demonstrating that a modality-perturbation curriculum within the proposed MAAF-Net framework significantly enhances robustness against missing sensor data without compromising clean accuracy or inference speed.
This paper introduces PREF, an interpretable vision-language model that successfully classifies political party affiliation from 2025 German election campaign media by verbalizing perceptual cues, achieving performance comparable to black-box embeddings while quantifying the extent to which party identity persists even after removing explicit branding.
This paper introduces a generalized Probabilistic Neutral Expressions (PNE) framework incorporating novel techniques like syncopation and carry-preserving to significantly improve key-recovery attacks on round-reduced ChaCha, achieving the first related-key attack on ChaCha8 and breaking 7.5 rounds of 256-bit ChaCha.
This paper proposes a novel memory-efficient Meet-in-the-Middle attack on round-reduced Ascon-Hash that leverages SAT modeling, Gaussian elimination, and triangulation to significantly reduce memory complexity while maintaining the best-known time complexities.
This paper proposes FedEDAuth, a lightweight federated learning framework that utilizes embedding-distribution analysis to detect and filter malicious clients, thereby achieving 100% detection of stealthy Byzantine data poisoning attacks and maintaining high accuracy in collaborative counterfeit integrated circuit detection.
This paper proposes a generalized LoD Transition Space framework that utilizes straight-skeleton-based 3D volumes to enable continuous, topologically consistent multi-scale transformations between diverse geometric primitives (polygons, polylines, and points), thereby supporting smooth map generalization across arbitrary Levels of Detail.