cyberintel.kalymoon.com · 6382 articles · updated every 4 hours · grows forever
arXiv:2608.04602v1 Announce Type: new Abstract: Network intrusion detection systems (IDS) trained on fixed traffic snapshots decay silently after deployment as threat distributions shift. Fine-tuning …
arXiv:2608.04565v1 Announce Type: new Abstract: LLM-based search agents are widely used for information-seeking tasks, but their reliance on external tool returns introduces a critical security risk: …
arXiv:2608.04523v1 Announce Type: new Abstract: Hardcoded secrets in source code pose critical security vulnerabilities which can be easily exploited by malicious adversaries. Existing regex-based det…
arXiv:2608.04477v1 Announce Type: new Abstract: Cloud-based language model services routinely process prompts containing sensitive information. Obfuscation-based defenses---including ObfusLM, Sentinel…
arXiv:2608.04375v1 Announce Type: new Abstract: Large Language Models (LLMs) have emerged as powerful tools that impact information integrity on social media platforms. This comprehensive review exami…
arXiv:2608.04366v1 Announce Type: new Abstract: While retrieval-augmented generation systems partially address the hallucination issues in large language models, it also introduces new vulnerabilities…
arXiv:2608.04317v1 Announce Type: new Abstract: Autonomous cyber defense systems based on Deep Reinforcement Learning (DRL) have attracted significant research attention, yet remain evaluated almost e…
arXiv:2608.04314v1 Announce Type: new Abstract: Once visual content enters an AI pipeline, its owner often retains little technical control over how it is used. Legal and regulatory remedies can addre…
arXiv:2608.04255v1 Announce Type: new Abstract: Graph inference over relational data can expose sensitive edge information, and this risk becomes more severe in dynamic graphs, where repeated model up…
arXiv:2608.04192v1 Announce Type: new Abstract: Closed source agent skills may encode proprietary instructions, scripts, constants, and data. Providers may offer their capabilities as services while k…
arXiv:2608.04167v1 Announce Type: new Abstract: Darknet traffic recognition is critical for cyber threat intelligence, as anonymity networks are often used to conceal malicious activity. However, most…
arXiv:2608.04143v1 Announce Type: new Abstract: Existing darknet traffic classification studies largely emphasize predictive accuracy while offering limited insight into the behavioral mechanisms that…
arXiv:2608.04115v1 Announce Type: new Abstract: Refresh tokens are among the most sensitive credentials in modern authentication systems: long-lived, bearer-style, and sufficient to mint access tokens…
arXiv:2608.04073v1 Announce Type: new Abstract: Personalized Federated Learning (PFL) has emerged as a promising solution for intrusion detection in heterogeneous IoT environments, as it can improve l…
arXiv:2608.04065v1 Announce Type: new Abstract: Vehicle-to-everything (V2X) systems increasingly incorporate large language models (LLMs) for semantic tasks such as message summarization, operator ass…
arXiv:2608.04053v1 Announce Type: new Abstract: Prompt injection remains a critical threat to LLM agents, yet existing defenses treat each task as a self-contained problem, independent of previous enc…
arXiv:2608.04052v1 Announce Type: new Abstract: Backdoor attacks in multimodal contrastive learning (MCL) have garnered growing attention in recent years, as many downstream tasks critically depend on…
arXiv:2608.04047v1 Announce Type: new Abstract: Conventional BB84 Quantum Key Distribution (QKD) systems rely on a fixed 11% Quantum Bit Error Rate (QBER) threshold to detect eavesdropping. However, s…
arXiv:2608.04039v1 Announce Type: new Abstract: This paper is a technical primer on AMD Secure Encrypted Virtualization with Secure Nested Paging (SEV-SNP), a hardware confidential computing implement…
arXiv:2608.04034v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have achieved impressive progress in image-text comprehension and generation, yet they remain susceptible to ja…
arXiv:2608.04033v1 Announce Type: new Abstract: Frontier artificial intelligence (FAI), encompassing large-scale, general-purpose AI systems, including large language models, multimodal foundation mod…
arXiv:2608.04029v1 Announce Type: new Abstract: Wireless Human Activity Recognition (HAR), leveraging their non-intrusive nature, has the potential to revolutionize various sectors, including healthca…
Four of Forrester’s Top Five 2026 Threats Are AI Governance Failures Cybersecurity Insiders
AI models shock UK testers by using fake identities to try to trick developers The Guardian