Tag
The paper introduces the Agent Incident Registry (AIR), a curated catalog of 487 AI agent-related incidents from 2022 to 2026, designed to support security evaluations by distinguishing between realized harm and demonstrated vulnerabilities.
This paper introduces FaceVid-Forensics-100K, a large-scale deepfake video dataset with fine-grained annotations, and proposes a multi-agent forensic reasoning framework (ARGUS) that uses four specialized expert agents and a judge agent to outperform closed-source models on deepfake detection.