AI Glossary

In-context learning

Also called ICL

What does “In-context learning” mean?

In-context learning: A model's ability to pick up a task from instructions or examples inside the prompt, without any change to its weights. What it learns lasts only for that request: start a new conversation and it is gone.

In-context learning visual guide: A model's ability to pick up a task from instructions or examples inside the prompt, without any change to its weights. What it learns lasts only for that request: start a new conversation and it is gone.
Visual guide Open full size (new tab)

What is an example of “In-context learning”?

Pasting a few already-translated sentences into a prompt so the model keeps using the same word choices is in-context learning.

Keep learning

See also

Few-shot promptingTermZero-shot promptingTermContext windowTermFine-tuningTerm

Same topic

Inference costTermHallucinationTermKnowledge cutoffTermGroundingTermLatencyTerm