2579 papers
This paper proposes AMH-APT, a novel APT detection framework that enhances provenance graph analysis by employing an adaptive dynamic masking strategy to preserve critical attack context and heterogeneity-aware projection to maintain distinct semantic spaces for diverse entity types, thereby outperforming existing methods in both log-level and entity-level detection tasks.
This study presents an offline, mobile-based diagnostic system utilizing a quantized MobileNetV2 model to achieve high-accuracy, real-time detection of soil-transmitted helminths and schistosomiasis from microscopic images, offering a cost-effective solution for point-of-care diagnosis in resource-constrained settings.
This study presents an intelligent predictive maintenance framework for military vehicles that utilizes a machine learning pipeline on engine health data, demonstrating that a Gradient Boosting model effectively enhances operational readiness by achieving high recall (97.4%) to minimize undetected engine failures while optimizing maintenance scheduling.
This paper presents and empirically validates an end-to-end Network Access Control architecture that integrates Active Directory, Microsoft NPS, and Cisco IOSvL2 switches to dynamically assign users to specific VLANs based on group membership via IEEE 802.1X authentication, demonstrating 100% success in valid authentication and effective isolation of unauthorized or failed attempts in a virtualized laboratory environment.
This paper proposes and evaluates a single, sector-agnostic AI framework using a Random Forest classifier on a 15-dimensional feature space that successfully detects both known and zero-day cyberattacks across five heterogeneous critical infrastructure domains while distinguishing them from physical faults, thereby demonstrating that a unified, explainable model can replace multiple sector-specific intrusion-detection systems.
This paper addresses the ethical opacity of multi-agent AI negotiations in hospital clinical decision support systems by synthesizing nine game-theoretic paradigms and proposing the novel Musical Chairs Coalition Convergence Game (MC3G) framework, which uses reputation-weighted convergence and mandatory human escalation to reduce silent failures and enhance accountability.
This paper demonstrates that GraphSAGE can achieve near-perfect accuracy in distinguishing prime from composite integers by learning structural differences in their modular-inverse graphs, whereas GCN fails to capture these distinctions due to its specific message-passing limitations.
This paper presents a benchmarking study evaluating five pre-trained seq2seq transformer models on the COSMMIC dataset across nine Indian languages to demonstrate that incorporating reader comments and image URLs alongside headlines and content significantly enhances zero-shot multimodal summarization performance.
This paper introduces RAIA, a risk-adaptive identity assurance framework that aggregates biometric continuity, QR session integrity, role-based authorization, and device supply-chain posture to dynamically manage access risks for IoT devices, demonstrating high attack containment and low benign friction in synthetic simulations while highlighting the critical need for deployment-specific threshold calibration.
This paper proposes a music style transfer framework utilizing conditional self-attention and hypersphere contrastive learning to effectively disentangle content and style representations, thereby generating high-quality, style-controllable musical outputs that support intelligent music education.