Edge computing and Internet of Things (IoT) have expanded the attack surface of modern networks. Security designs often tradeoff detection quality and privacy: centralized trust creates single points of failure, while distributed approaches may sacrifice accuracy or formal privacy guarantees. This article presents a federated trust modeling framework that integrates multimodal anomaly detection, Byzantine-resilient federated learning with $(\epsilon,\delta)$-differential privacy (DP), and context-aware zero trust architecture decision-making. The three-layer architecture comprises device-level trust learning with enhanced variational autoencoder, Isolation Forest, long short-term memory, and statistical process control modalities; a federated aggregation layer for robust aggregation with DP; and a trust scoring and decision layer that maps evidence to continuous trust and access levels using a subjective logic-inspired formulation. We provide detailed algorithmic implementations with pseudocode for each layer. Comprehensive evaluation demonstrates exceptional performance: high precision with low false positive rate, near-linear scalability achieving high efficiency, high accuracy with precision detecting most of attacks with zero false alarms, and sublinear time complexity. Privacy preservation is maintained through DP guarantees without accuracy degradation. The results support deployability studies for large-scale IoT and Artificial Intelligence of Things settings, with generalization to real telemetry left to future work.
Shengjie Xu, Yi Qian· IEEE Journal of Selected Are...· 0 citations
This paper provides a tutorial and survey of LLM-based agents for automated vulnerability discovery and penetration testing. The rapid advancements in Large Language Models (LLMs) have opened new possibilities for their application in cybersecurity. Unlike traditional tools, LLMs can process natural language, retain context, and perform structured reasoning, enabling automation of complex security workflows. By equipping these models with memory mechanisms and tool-use interfaces, researchers have developed LLM-driven autonomous agents capable of performing specialized tasks such as web application testing, source code auditing, fuzzing, and network security assessment. Multi-agent architectures have emerged in which agents assume specialized roles (e.g., distinct penetration-test phases or exploit types). In addition to surveying existing applications, this work provides a step-by-step walkthrough of integrating agentic AI into penetration testing workflows. The walkthrough explores four phases: (I) mission scoping and prompt engineering for test definition and constraint enforcement, (II) autonomous exploration and tool selection for target interaction, (III) vulnerability hypothesis formation and verification through experiment design and feedback, and (IV) payload generation and refinement to transform validated findings into concrete exploits. The survey synthesizes 155 cited sources (2022–early 2026) using explicit inclusion/exclusion criteria, with coverage extending to reasoning-class models, the Model Context Protocol ecosystem, autonomous bug-bounty agents, and standardized agentic security benchmarks. Despite progress, challenges persist, such as model hallucination, context limitations, operating costs, and ethical concerns. Future directions include advancements in multi-agent collaboration, mechanisms for agent self-improvement, and improved transparency into agent behavior.
Robel Addis, Kiran Porter, J. Ryan et al.· IEEE Access· 0 citations
