In-Context Learning
Definition
In-context learning means that a model uses examples within the current input.
The ability of a language model to learn new tasks or formats directly from examples provided promptly, without changing the model weights.
Simply explained
Like learning on the model sheet for this one task without retraining the model.
Why it matters in practice
Helps to realistically classify the possibilities, quality and limitations of AI systems.
Example from business
Possible example: Three well-tested example answers show an assistant the desired format.
Limits and pitfalls
- Keep the examples of permanent training or use sensitive sample data.
Origin & history
Brown and co-authors examined 2020 Few-Shot performances of GPT-3 using examples in context, without task-specific weight updating.
Use in mid-sized companies
- Possible application in SMEs: Three well-tested example answers show an assistant the desired format.
Frequently asked questions
What does learning mean in-context learning?
In-context learning means that a model uses examples within the current input.
What errors should be avoided?
Keep the examples of permanent training or use sensitive sample data.
Sources
- Language Models are Few-Shot Learners · Brown et al. (2020), Language Models are Few-Shot Learners
Sources checked on 09.10.2026. Editorial use cases are hypothetical; no evidence of real customer results.
Editorially reviewed on: