Skip to content

Model Training & Adaptation

Zero-shot learning

AI handling a task or recognizing a group without having been taught labeled examples of that specific target.

Example

An AI system recognizes a new category using its description.

Why people use it

It lets AI attempt unfamiliar categories using prior knowledge or descriptions.

What you'll hear

“Can it recognize this category without labeled examples of it?”

What this means for you

Check what prior knowledge and auxiliary information the AI system used.

Can you control it?

Developer-only

The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.

Common questions

Is zero-shot learning identical to giving a prompt without examples?
No. Zero-shot prompting is a prompting setup; zero-shot learning is a broader capability setting.
Does zero-shot mean it has never learned anything relevant?
No. It usually relies on earlier learning or information connecting the new task with familiar concepts.
Can a more helpful description improve the result?
It can, when the method uses descriptions to understand the new category.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice