US Patent Issued to NAVER on March 31 for "Method and system for fine-tuning neural conditional language models using constraints" (French, Polish Inventors)

ALEXANDRIA, Va., March 31 -- United States Patent no. 12,591,745, issued on March 31, was assigned to NAVER Corp. (Seongnam-si, South Korea).

"Method and system for fine-tuning neural conditional language models using constraints" was invented by Tomasz Korbak (Warsaw, Poland), Hady Elsahar (Grenoble, France), German Kruszewski (Saint Mande, France) and Marc Dymetman (Grenoble, France).

According to the abstract* released by the U.S. Patent & Trademark Office: "A processor-implemented method for fine-tuning a pre-trained neural conditional language model to perform a downstream task. A pre-trained conditional language model and at least one target constraint for satisfying a task-related control objective are received. A neural model is t...