Describe zero-shot chain-of-thought and how it compares to standard chain-of-thought prompting.

• A. Zero-shot CoT asks for step-by-step reasoning without examples; standard CoT uses few-shot demonstrations that include reasoning steps.
• B. Zero-shot CoT requires many in-context examples; standard CoT uses none.
• C. Zero-shot CoT never asks for reasoning.
• D. Zero-shot CoT uses external tools to compute steps.

Answer: (D)

Zero-shot chain-of-thought asks the model to produce a step-by-step solution without showing any worked examples. You prompt it to think aloud or reason through the problem on its own, often with wording like “Think step by step,” but there are no prior demonstrations for the model to imitate.

Standard chain-of-thought prompting, by contrast, uses few-shot demonstrations: you provide several examples that include both the question and the full reasoning steps leading to the answer. The model then imitates that reasoning pattern in new problems. This explicit exposure to how to reason helps the model structure its solution more carefully and often improves accuracy on complex tasks.

The key distinction is whether reasoning demonstrations are included in the prompt. External tools to compute steps aren’t what defines zero-shot CoT; tool use is a separate capability and isn’t required for chain-of-thought prompting.