Back to the Index

Prompt Tuning

What is Prompt Tuning?

Prompt tuning (PT) is a parameter-efficient adaptation method for LLMs that adds a small number of tunable embeddings to an otherwise frozen model. Prompts do not have to be written in natural language (e.g., English). In prompt tuning (or prefix-tuning), floating-point-valued vectors are searched directly by gradient descent, to maximize the log-probability of outputs.

Does this content look outdated? If you are interested in helping us maintain this, feel free to contact us.

© Hopsworks 2024. All rights reserved. Various trademarks held by their respective owners.

Privacy Policy
Cookie Policy
Terms and Conditions