Few-Shot Learning

Putting examples in the prompt instead of training on them.

technique Prompting 4 relationships

Aperçu

Few-shot learning supplies a handful of worked examples in the prompt so the model infers the task from them without any weight update. It solves the problem that fine-tuning needs labelled data, a training run and a deployment per task, while many tasks are simple enough to demonstrate in a paragraph. It exists because it emerged from scale rather than design: the capability appeared in large models that were never trained for it.

Graphe des technologies

Comment Few-Shot Learning se rattache au reste de la pile.

Utilise et dépend de

LLM concept Models Few-Shot Learning based on LLM High ✓ source

Appliqué à

Prompt Engineering technique Prompting Few-Shot Learning used for Prompt Engineering High ✓ source

Outils

Logiciels qui implémentent Few-Shot Learning ou s'y intègrent.

Prompting

In-Context Learning

Alternatives

Zero-Shot concept
Une alternative à Few-Shot Learning · source

Comparaisons

Comparaisons détaillées avec les alternatives ci-dessus.

Few-Shot Learning vs Zero-Shot

Position de la technologie

AI
↓
Prompting
↓
technique
↓
Few-Shot Learning

Technologies liées

In-Context Learning concept
LLM concept
Prompt Engineering technique
Zero-Shot concept

Ressources officielles

Documentation

Dernière mise à jour: 2026-10-08
Dernière vérification: 2026-10-08

Sources citées

Chaque relation de cette page porte sa source.

arxiv.org 4 claims