arXiv Brief — 2026-10-09
New papers on Common Criteria evaluation failures, LLM coding agent skills, and MoE router telemetry privacy.
Llm-Security
Conference Brief — 2026-10-08
New papers on botnet remediation, satellite honeypots, Payment Channel Networks, censorship circumvention, hypervisor fuzzing, and LLM prompt leakage.
arXiv Brief — 2026-10-07
Today’s research explores self-propagating misalignment in LLMs, rigorous behavioral watermarking for agents, and dual intrusion detection for vehicular networks.
arXiv Brief — 2026-10-06
An attack injecting vulnerabilities via AI coding assistants, task-level poisoning detection in instruction-tuned LLMs, and fingerprinting black-box dense retrievers.
Conference Brief — 2026-10-06
A batch of papers from NDSS.