Patents Assigned to Character Technologies Inc.
  • Optimizing key value cache for large language model inference
    Patent number: 12682211
    Abstract: An input sequence is received from a client device. Large language model inference is performed by processing the input sequence through a series of transformer layers to generate one or more tokens including by performing hybrid attention, multi-query attention, and cross-layer key value sharing. The one or more generated tokens are provided to the client device.
    Type: Grant
    Filed: April 28, 2025
    Date of Patent: July 14, 2026
    Assignee: Character Technologies Inc.
    Inventors: Bowen Liang, Noam Mordechai Shazeer, Myle Ott
  • Retraining a large language model
    Patent number: 12664375
    Abstract: An indication of a problem associated with a response generated by a first artificial intelligence (AI) language model is received. The first AI language model generates one or more modified responses for the response. The one or more modified responses and some or all of a turn-based conversation that included the response are provided to a second AI language model. The first AI language model is retrained based in part on an output of the second AI language model.
    Type: Grant
    Filed: May 14, 2025
    Date of Patent: June 23, 2026
    Assignee: Character Technologies Inc.
    Inventors: Satyaki Chakraborty, Landon Jonquiere Brittan Smith, Te-Lin Wu
  • Character turn summaries
    Patent number: 12481835
    Abstract: A history of conversation turns is obtained. A prompt and the history of conversation turns is provided to an artificial intelligence language model. The history of conversation turns includes one or more summaries summarizing a plurality of the conversation turns. A response is received from the AI language model.
    Type: Grant
    Filed: March 19, 2025
    Date of Patent: November 25, 2025
    Assignee: Character Technologies Inc.
    Inventors: Denis Titoruk, Jonathan Michael Abrams, James Douglas Groeneveld
  • Audio turn understanding system
    Patent number: 12322384
    Abstract: A real-time audio stream associated with a user is segmented into one or more chunks of audio. The one or more segmented chunks of audio are provided to an audio understanding model. It is determined that the user is finished with their turn in a conversation. In response to determining that the user has finished with their turn in the conversation, a response is provided based on the real-time audio stream.
    Type: Grant
    Filed: September 27, 2024
    Date of Patent: June 3, 2025
    Assignee: Character Technologies Inc.
    Inventors: Prajit Ramachandran, Yi Cui