ALEXANDRIA, Va., Aug. 25 -- United States Patent no. 12,718,030, issued on Aug. 25, was assigned to Amazon Technologies Inc. (Seattle).

"Large language models providing evidence mappings for generated output" was invented by Saket Dingliwal (Kirkland, Wash.), Karthik Gopalakrishnan (Sunnyvale, Calif.), Sravan Babu Bodapati (Fremont, Calif.), Sarthak Handa (Seattle) and Katrin Kirchhoff (Seattle).

According to the abstract* released by the U.S. Patent & Trademark Office: "A data set comprising a plurality of records is obtained. A given record of the data set includes a source set of text tokens to be analyzed, and a corresponding annotated analysis result. An annotation corresponding to a text sequence of the result indicates a portion of...