Build in-context RL for skill internalization, improving agent learning on ALFWorld and Search-QA with fewer updates
Official code for "SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization"