Toward Efficient Agents: Memory, Tool learning, and Planning - Explained Simply | ArXiv Explained