Which prompting technique asks the model to think step by step to solve complex reasoning tasks?

Which prompting technique asks the model to think step by step to solve complex reasoning tasks?

• A. Chain of Thought prompting
• B. Prompt Engineering
• C. Context Length
• D. Temperature

Answer: (A)

Chain-of-Thought prompting is the technique that asks the model to think step by step to solve complex reasoning tasks. By prompting the model to produce intermediate reasoning steps before giving the final answer, you encourage a structured approach to solving problems that require multi-step calculations, logical deduction, or planning. This helps the model reveal its reasoning path and often leads to more accurate results because it reduces the tendency to jump to a guess without examining all parts of the problem. The other terms don’t target step-by-step reasoning in the same way. Prompt engineering is a general practice of crafting prompts to shape outputs, not specifically about eliciting explicit chain-of-thought. Context length refers to how much input the model can consider at once, and temperature controls how random or deterministic the output is; neither directly prompts the model to show its reasoning steps.

Chain-of-Thought prompting is the technique that asks the model to think step by step to solve complex reasoning tasks. By prompting the model to produce intermediate reasoning steps before giving the final answer, you encourage a structured approach to solving problems that require multi-step calculations, logical deduction, or planning. This helps the model reveal its reasoning path and often leads to more accurate results because it reduces the tendency to jump to a guess without examining all parts of the problem.

The other terms don’t target step-by-step reasoning in the same way. Prompt engineering is a general practice of crafting prompts to shape outputs, not specifically about eliciting explicit chain-of-thought. Context length refers to how much input the model can consider at once, and temperature controls how random or deterministic the output is; neither directly prompts the model to show its reasoning steps.