ALEXANDRIA, Va., Sept. 24 -- United States Patent no. 12,743,459, issued on Sept. 22, was assigned to JPMORGAN CHASE BANK N.A. (New York).
"Method and system for optimizing a chunk size for a large language model" was invented by Sidharth Talwar (Greater Noida, India), Vipin Agarwal (Bangalore, India), Indu Indiramani (Bangalore, India), Sanjay Saran Garg (Bangalore, India) and Ranjit Radhakrishnan (Mumbai, India).
According to the abstract* released by the U.S. Patent & Trademark Office: "A method and system for optimizing a chunk size for a large language model (LLM) are disclosed. The method includes receiving at least one from among a set of input data and a set of evaluation metrics for evaluating LLM performance. Next, the method in...