US Patent Issued to GOOGLE on July 14 for "Blockwise controlled decoding of natural language (NL) based output generated using a large language model (LLM) to reduce latency in rendering thereof" (California, New York Inventors)

Posted On: 2026-07-14

ALEXANDRIA, Va., July 14 -- United States Patent no. 12,681,964, issued on July 14, was assigned to GOOGLE LLC (Mountain View, Calif.).

"Blockwise controlled decoding of natural language (NL) based output generated using a large language model (LLM) to reduce latency in rendering thereof" was invented by Sidharth Mudgal (Mountain View, Calif.), Ahmad Beirami (New York), Jilin Chen (Cupertino, Calif.), Alex Beutel (Brooklyn, N.Y.), Harish Ganapathy (Sunnyvale, Calif.), YaGuang Li (Sunnyvale, Calif.), Tao Wang (Sunnyvale, Calif.), Yanping Huang (Mountain View, Calif.) and Trevor Strohman (Sunnyvale, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Implementations relate to reducing latency in generating a...

Click here to read full article from source

To read the full article or to get the complete feed from this publication, please Contact Us.

Exclusive

Category

Source

Publication

Location

US Patent Issued to GOOGLE on July 14 for "Blockwise controlled decoding of natural language (NL) based output generated using a large language model (LLM) to reduce latency in rendering thereof" (California, New York Inventors)