Vulnerability-Discovery

New papers on agents as operating systems, LLM-based vulnerability detection, RAG poisoning, LLM agent root cause analysis harnesses, and agentic message boards.

New attacks against aligned LLMs, automatic exploit generation for Node.js, mapping CVEs to MITRE ATT&CK, and cross-privilege BTB side channels.

A batch of NDSS papers covering URL shortening service abuse, GPU driver fuzzing, ZK compiler testing, and latency improvements for DAG-based consensus.

New work from NDSS 2026, including intent fuzzing, container desynchronization risks, and memory backdoor attacks on neural networks.

Browser agent confused deputy vulnerabilities, LLM retrieval access controls, and new defense benchmarks.