US Patent Application 17718738. Parameter Efficient Prompt Tuning for Efficient Models at Scale simplified abstract
Contents
Parameter Efficient Prompt Tuning for Efficient Models at Scale
Organization Name
Inventor(s)
Brian David Lester of Mountain View CA (US)
Rami Al-rfou of Menlo Park CA (US)
Noah Constant of Los Angeles CA (US)
Parameter Efficient Prompt Tuning for Efficient Models at Scale - A simplified explanation of the abstract
- This abstract for appeared for US patent application number 17718738 Titled 'Parameter Efficient Prompt Tuning for Efficient Models at Scale'
Simplified Explanation
This abstract describes a method for natural language processing that uses trained prompts to guide a pre-trained machine learning model to generate specific outputs for a given task. The approach involves training a subset of parameters for the task and then inputting them along with input data into the pre-trained model. The pre-trained model's parameters are frozen during the prompt training, which helps reduce computational resources while still benefiting from the knowledge learned by the pre-trained model.
Original Abstract Submitted
Systems and methods for natural language processing can leverage trained prompts to condition a large pre-trained machine-learned model to generate an output for a specific task. For example, a subset of parameters may be trained for the particular task to then be input with a set of input data into the pre-trained machine-learned model to generate the task-specific output. During the training of the prompt, the parameters of the pre-trained machine-learned model can be frozen, which can reduce the computational resources used during training while still leveraging the previously learned data from the pre-trained machine-learned model.
