US Patent Issued to Electronics and Telecommunications Research Institute on Dec. 2 for "Method of domain-adapting large-capacity pre-trained language model using semantic chunk dynamic weight masking" (South Korean Inventors)

ALEXANDRIA, Va., Dec. 2 -- United States Patent no. 12,488,187, issued on Dec. 2, was assigned to Electronics and Telecommunications Research Institute (Daejeon, South Korea).

"Method of domain-adapting large-capacity pre-trained language model using semantic chunk dynamic weight masking" was invented by Euisok Chung (Daejeon, South Korea), Hyun Woo Kim (Daejeon, South Korea), Hwajeon Song (Daejeon, South Korea), Jeongmin Yang (Daejeon, South Korea), Byunghyun Yoo (Daejeon, South Korea) and Ran Han (Daejeon, South Korea).

According to the abstract* released by the U.S. Patent & Trademark Office: "A domain adaptation procedure, such as fine-tuning training, is required to utilize a large-capacity PLM for a specific domain. Attempts in exis...