Chain-of-Thought Prompting Elicits Reasoning in Large Language Models - Explained Simply | ArXiv Explained