ModelRefs / In-Context Learning (ICL) — AI Glossary
In-Context Learning (ICL) — AI Glossary
The ability of large language models to perform new tasks from examples provided in the prompt, without any weight updates.
Overview
ICL emerges at scale: the model infers a task from demonstration examples in the context window and applies it to a new input. It enables rapid task adaptation without fine-tuning. Performance depends on example quality, format, and ordering. Related to but distinct from few-shot prompting—ICL is the mechanism, few-shot is the technique.
Reference details
| Topic | architecture |
|---|---|
| Also known as | ICL, few-shot learning |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to In-Context Learning (ICL) — AI Glossary.
Frequently asked questions
What is In-Context Learning (ICL)?
The ability of large language models to perform new tasks from examples provided in the prompt, without any weight updates.
Is In-Context Learning (ICL) the same as ICL?
Yes — ICL, few-shot learning are common aliases for In-Context Learning (ICL).
What concepts are related to In-Context Learning (ICL)?
Closely related concepts include few shot, zero shot, emergence, context window.