Q4
Single choice
A support operations team wants an LLM to generate multi-step troubleshooting guides by progressively refining the reasoning chain.
Which prompting approach should they adopt?