In-Context Learning

term_id: in_context_learning

Category: application_paradigms

Definition

In-context learning (ICL) allows large language models to adapt to new tasks without updating their weights. By providing input-output pairs within the prompt context, the model infers the pattern and applies it to new queries. This zero-shot or few-shot capability enables rapid prototyping and flexibility, serving as a powerful alternative to traditional fine-tuning for tasks requiring quick adaptation to novel domains.

Summary

A technique where models learn to perform tasks by observing examples provided in the prompt.

Key Concepts

  • Few-Shot Learning
  • Zero-Shot
  • Prompt Design
  • Weight-Free Adaptation

Use Cases

  • Quickly testing model capabilities on new datasets
  • Dynamic task switching in conversational agents
  • Prototyping AI applications without retraining

Code Example

1
2

prompt = "Translate English to French:\nEnglish: Hello\nFrench: Bonjour\nEnglish: Cat\nFrench:"
response = model.generate(prompt)

Related