Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents Paper • 2605.25971 • Published May 25 • 16
WorldReasonBench: Human-Aligned Stress Testing of Video Generators as Future World-State Predictors Paper • 2605.10434 • Published May 11 • 29
Dynamic Skill Lifecycle Management for Agentic Reinforcement Learning Paper • 2605.10923 • Published May 11 • 13
CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability Paper • 2602.03012 • Published Feb 3 • 3
CoBA-RL: Capability-Oriented Budget Allocation for Reinforcement Learning in LLMs Paper • 2602.03048 • Published Feb 3 • 32