AI RESEARCH
RouteGuard: Internal-Signal Detection of Skill Poisoning in LLM Agents
arXiv CS.AI
•
ArXi:2604.22888v1 Announce Type: cross