CCAR-P

Question

You are assessing prompting claims in a peer’s design document.

For each of the following statements, select Yes if the statement is true. Otherwise, select No.

Yes or No

Statements	Yes	No
Zero-shot is appropriate as a starting point for closed-set classification when categories and descriptions are provided in the prompt.
Adding a few-shot example set with exact field names anchors output formatting on extraction tasks.
Chain-of-thought should be the default on every task regardless of complexity.
Leading-question framing is best mitigated by asking for an even-handed comparison with stated criteria.
Behavioral differences across model versions can be ignored once a prompt has been validated on one version.

Explanation

Zero-shot classification works when labels and their definitions are clearly specified. Few-shot examples with exact schema fields reinforce the desired extraction format. Chain-of-thought should be applied only when useful for the task, not as a blanket default. Neutral comparisons with explicit criteria counter leading framing. Model behavior can vary by version, so prompts require validation after version changes.

CCAR-P

AstroTutor

Notes

Readiness

QuestionQ33

Community Discussion