US Patent Issued to BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY on Nov. 18 for "Method for training non-autoregressive translation model" (Chinese Inventors)

ALEXANDRIA, Va., Nov. 18 -- United States Patent no. 12,475,328, issued on Nov. 18, was assigned to BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY Co. LTD. (Beijing).

"Method for training non-autoregressive translation model" was invented by Xiyang Wang (Beijing), Ruiqing Zhang (Beijing), Zhongjun He (Beijing), Zhi Li (Beijing) and Hua Wu (Beijing).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method for training a non-autoregressive translation (NAT) model includes: acquiring a source language text, a target language text corresponding to the source language text and a target length of the target language text; generating a target language prediction text and a prediction length by inputting the source language ...