WebWorld: The Browser as a World Model for Self-Improving Web Code Paper • 2608.30530 • Published 24 days ago • 10
EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning Paper • 2608.06197 • Published Aug 6 • 47
Self-State Attacks on Self-Hosted AI Agents: How Far Can OS Defenses Go? Paper • 2607.17986 • Published Jul 20 • 6
LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget Paper • 2607.14952 • Published Jul 16 • 91