ALEXANDRIA, Va., Oct. 28 -- United States Patent no. 12,456,013, issued on Oct. 28, was assigned to Salesforce Inc. (San Francisco).
"Systems and methods for training a neural network model using knowledge from pre-trained large language models" was invented by Shiva Kumar Pentyala (Mountain View, Calif.), Prafulla Kumar Choubey (San Jose, Calif.), Shashank Harinath (Mountain View, Calif.), Sitaram Asur (Newark, Calif.), Chien-Sheng Jason Wu (Mountain View, Calif.), Zachary Alexander (Berkeley, Calif.) and Caiming Xiong (Menlo Park, Calif.).
According to the abstract* released by the U.S. Patent & Trademark Office: "Embodiments described herein provide a training framework for generative NLP models that operate on previously learnt knowle...