AI Glossary Term

What does “In-Context Learning” mean?

Definition

A model's ability to follow patterns from examples provided directly in the prompt.

Browse the full AI Glossary Explore all free AI guides

How In-Context Learning works

Example

You paste three examples of converting messy dates to ISO format and then a fourth messy date. The model returns it correctly formatted, even though it was never retrained for that task.

Why it matters

It lets non-experts adapt models just by writing prompts. The adaptation lasts only for that conversation.

Common questions

Does the model really learn?

Its weights stay unchanged. It uses the examples as context during that request, so nothing is kept afterward.

How is it related to few-shot learning?

Few-shot prompting is the most common way to use in-context learning.

Does it always work?

No. Performance depends on the model, the task, and how clear and relevant the examples are.

PromptThe input instructions and context provided to a generative model.Few-Shot LearningLearning or adapting behavior from only a small number of examples.Machine Learning (ML)Methods that allow systems to learn patterns from data and improve over time.Unsupervised LearningLearning patterns from unlabeled data without explicit target outputs.SentencePieceA language-agnostic tokenizer that learns subword units directly from raw text without pre-splitting on whitespace.Intent ClassificationPredicting the user's purpose from a text query to route it correctly.

Learn more in our free guides

Deep Learning Reinforcement Learning Machine Learning Basics Supervised Learning

Test yourself

Deep Learning Quiz Reinforcement Learning Quiz Machine Learning Basics Quiz