Learn Before
Concept

Sufficiency of Demonstrations in Few-Shot Learning

For a large language model to successfully perform a task via few-shot learning, the prompt must contain a sufficient number of demonstrations. Although the required number of examples is generally small, it must be adequate for the model to recognize the underlying input-output pattern of the task.

0

1

Updated 2026-04-29

Contributors are:

Who are from:

Tags

Ch.3 Prompting - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences