Patents by Inventor Xiaokai Wei

Xiaokai Wei has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Large language models utilizing element-wise operation-fusion
    Patent number: 12346673
    Abstract: Techniques for using a quantized and/or fused model are described. In some examples, a service is to receive a request to use a trained model, the request including input data; apply a trained model to the input data, the application of the trained model includes fusing one or more matrix multiplication operations with element-wise operations; and output a result from the trained model.
    Type: Grant
    Filed: March 31, 2023
    Date of Patent: July 1, 2025
    Assignee: Amazon Technologies, Inc.
    Inventors: Sujan Kumar Gonugondla, Bohan Yao, Haifeng Qian, Xiaokai Wei, Jiacheng Guo, Vamshidhar Krishnamurthy Dantu, Praphruetpong Athiwaratkun, Ramesh M. Nallapati, Parminder Bhatia, Srinivas Iragavarapu, Yuchen Tian, Rama Krishna Sandeep Pokkunuri, Sudipta Sengupta, Bing Xiang
  • Systems and methods for classifying content items based on dynamic signals
    Patent number: 11250030
    Abstract: Systems, methods, and non-transitory computer readable media can obtain dynamic signals associated with content items. A machine learning model can be trained based on the dynamic signals associated with the content items. A classification for a content item can be determined based on the machine learning model.
    Type: Grant
    Filed: May 2, 2018
    Date of Patent: February 15, 2022
    Assignee: Facebook, Inc.
    Inventors: Xiaokai Wei, Eric Hsin-Chun Huang
  • SYSTEMS AND METHODS FOR CLASSIFYING CONTENT ITEMS BASED ON DYNAMIC SIGNALS
    Publication number: 20190340293
    Abstract: Systems, methods, and non-transitory computer readable media can obtain dynamic signals associated with content items. A machine learning model can be trained based on the dynamic signals associated with the content items. A classification for a content item can be determined based on the machine learning model.
    Type: Application
    Filed: May 2, 2018
    Publication date: November 7, 2019
    Inventors: Xiaokai Wei, Eric Hsin-Chun Huang