Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning

ArXi:2605.28424v1 Announce Type: new Equipping large language models with explicit skills has emerged as a promising paradigm for enabling autonomous agents to solve complex tasks. Agent skills can be inherently divided into general skills for broad cognitive transfer and task-specific skills for dynamic execution. However, existing skill-based reinforcement learning (RL) methods typically force a rigid choice between full externalization, which incurs prohibitive context overhead, and full internalization, which risks overfitting and knowledge conflicts.