ALEXANDRIA, Va., Dec. 31 -- United States Patent no. 12,511,479, issued on Dec. 30, was assigned to Salesforce Inc. (San Francisco).
"Systems and methods for in-context learning using small-scale language models" was invented by Hailin Chen (Singapore), Shafiq Rayhan Joty (San Jose, Calif.), Amrita Saha (Singapore) and Chu Hong Hoi (Singapore).
According to the abstract* released by the U.S. Patent & Trademark Office: "Embodiments described herein provide a data driven framework that (i) translates demonstration examples to a fixed-length soft prompt-a sequence of soft tokens; and (ii) learns a global (not generated from demonstrations) soft prompt. The framework then combines the global prompt, the translated prompts and the original con...