Few-Shot

Definitions

1. Describing a setup where a model is given a small number of examples in the prompt to demonstrate the desired behavior before handling a new input. Few-shot prompting is used to steer output format, task framing, or style without changing the underlying model weights.

   In plain English: Giving an AI a few examples first so it can follow the pattern.

   Example: "The classifier became much more consistent once they switched from a zero-shot prompt to a few-shot prompt with three labeled examples."

Related Terms

• Chain of Density
• Chain of Verification
• Context Stuffing
• Interactive Learning
• Passive Learning
• Self-Supervised
• Soft Prompt
• Supervised
• Unlocking
• Zero-Shot