Search for a command to run...
Beyond Reasoning: Reinforcement Learning Unlocks Parametric Knowledge in LLMs