few-shot prompting
Few-shot prompting provides worked examples of the desired task behavior.
Use this when A natural-language instruction leaves the expected output format ambiguous.
without the term
“The AI keeps formatting the output slightly differently each time. Can I just show it a couple of examples of exactly what I want so it copies the pattern?”
with the term · How do you ask a model for few-shot prompting?
Add a few representative examples for this output contract. Evaluate on separate cases, including one that should be rejected.
What is few-shot prompting?
The prompt includes a few worked input and output examples. They must be consistent and cover the distinctions the model needs, because the model copies the examples more than the instructions.
Why does few-shot prompting matter when vibe coding?
Inconsistent examples can teach the wrong behavior more strongly than the instruction.
How do you do few-shot prompting?
Use a small consistent set and evaluate on different inputs.
What goes wrong with few-shot prompting?
Reusing the demonstrations as the entire evaluation set.