HazardAuditor: From Executable Threats to Safer Computer-Use Agents Paper • 2609.15134 • Published 10 days ago • 18
WithEveryone: Unified Planning and Identity Grounding for Group Image Generation Paper • 2608.20336 • Published Aug 20 • 42
HarmProfile: Characterizing Harmful Distributions in Frontier LLMs Paper • 2608.14577 • Published Jun 11 • 20