Glossary term
Glossary term
Agentic Systems
A technique for tuning a large language model for a particular task, without resource intensive fine-tuning. Instead of retraining all the weights in the model, soft prompt tuning automatically adjusts a prompt to achieve the same goal.
Given a textual prompt, soft prompt tuning typically appends additional token embeddings to the prompt and uses backpropagation to optimize the input.
A "hard" prompt contains actual tokens instead of token embeddings.
Created for this library
A SaaS team uses soft prompt tuning to specialize a base LLM per customer with a small number of trainable parameters.
A research team uses soft prompt tuning to adapt a foundation model to several domains while keeping the base weights shared.
An enterprise legal team uses soft prompt tuning to specialize a base model on contract review without modifying the base weights.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License