Skip to content

Model Training & Adaptation

Prompt tuning

Teaching a set of numerical instructions to guide an AI while leaving most of its learning unchanged.

Example

A training process learns a soft prompt that helps an AI system classify text.

Why people use it

It adapts AI to a task by learning a small set of guiding values.

What you'll hear

“This is learned guidance, not just a rewritten sentence.”

What this means for you

Distinguish AI system adaptation methods from manual prompt editing.

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 prompt tuning just rewriting instructions by hand?
No. Soft prompt learned settings are learned numerically rather than simply edited as ordinary text.
Will the learned prompt look like ordinary writing?
Often not. It can be a set of numbers that the AI uses internally.
Can it work equally well for every task?
No. The task, teaching examples and original AI system affect the result.

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