Contrast zero-shot and few-shot prompting. When is few-shot the simplest reliable win?
LLM & AILevel: IntermediatePrompting & In-Context Learning
Contrast zero-shot and few-shot prompting. When is few-shot the simplest reliable win?
Hints
- Identify the stage: pretraining, adaptation, generation or evaluation.
- Name the inputs, constraints and a case where the approach fails.
Answer
Zero-shot gives the model only the instruction and trusts it to comply. Few-shot pastes a handful of worked examples first, so the model copies the pattern. Few-shot is the simplest dependable boost whenever the exact format, style, or labelling matters.
Source and licence: Skillumen — Awesome AI Engineer Interview Questions · CC0-1.0