Few-Shot Learning

Resumen

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.

Grafo de tecnologías

Cómo se conecta Few-Shot Learning con el resto del stack.

Herramientas

Software que implementa o se integra con Few-Shot Learning.

Alternativas

• Zero-Shot concept

  Una alternativa a Few-Shot Learning · source

Comparativas

Comparativas detalladas con las alternativas anteriores.

• Few-Shot Learning vs Zero-Shot