Skip to content

Generative AI & LLMs

Parameter-efficient fine-tuning

Adapting an AI by changing a relatively small part of its learned settings.

Example

A team adjusts small additions to an AI while leaving most of its earlier learning unchanged.

Why people use it

It reduces some of the time and memory needed to adapt a large AI system.

What you'll hear

“Can we teach the new task without changing everything?”

What this means for you

Compare the adapted system with the original on the jobs you need.

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

Does learned setting-efficient mean the result is always better?
No. It reduces some training demands but performance still depends on the task.
Does it remove the need for teaching examples?
No. The adaptation still needs suitable information and a clear learning task.
Can the adapted part be shared separately?
Often. Some methods allow a smaller learned addition to be stored separately from the original system.

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