US Patent Issued to Microsoft Technology Licensing on Nov. 25 for "Pre-training a unified natural language model with corrupted span and replaced token detection" (Washington Inventors)

ALEXANDRIA, Va., Nov. 25 -- United States Patent no. 12,481,841, issued on Nov. 25, was assigned to Microsoft Technology Licensing LLC (Redmond, Wash.).

"Pre-training a unified natural language model with corrupted span and replaced token detection" was invented by Pengcheng He (Sammamish, Wash.), Jianfeng Gao (Woodinville, Wash.), Nanshan Zeng (Bellevue, Wash.), Xuedong Huang (Yarrow Point, Wash.), Wei Xiong (Bellevue, Wash.) and Baolin Peng (Issaquah, Wash.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Systems and methods are provided for training and using a novel unified language foundation model. An encoder-decoder natural language model is obtained and various training data is obtained and used for tr...