中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

14 records (took 0.012 seconds)

    GENERATING NEURAL NETWORK OUTPUTS USING INSERTION COMMANDS
    1.
    发明申请
    GENERATING NEURAL NETWORK OUTPUTS USING INSERTION COMMANDS 审中-公开

    公开(公告)号：US20200372356A1

    公开(公告)日：2020-11-26

    申请号：US16883772

    申请日：2020-05-26

    Applicant: Google LLC

    Inventor： William Chan ,  Mitchell Thomas Stern ,  Nikita Kitaev ,  Kelvin Gu ,  Jakob D. Uszkoreit

    IPC: G06N3/08 ,  G06N3/04 ,  G06F40/237

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing sequence modeling tasks using insertions. One of the methods includes receiving a system input that includes one or more source elements from a source sequence and zero or more target elements from a target sequence, wherein each source element is selected from a vocabulary of source elements and wherein each target element is selected from a vocabulary of target elements; generating a partial concatenated sequence that includes the one or more source elements from the source sequence and the zero or more target elements from the target sequence, wherein the source and target elements arranged in the partial concatenated sequence according to a combined order; and generating a final concatenated sequence that includes a finalized source sequence and a finalized target sequence, wherein the finalized target sequence includes one or more target elements.

    Parallel decoding using autoregressive machine learning models
    2.
    发明授权
    Parallel decoding using autoregressive machine learning models 有权

    公开(公告)号：US10521701B2

    公开(公告)日：2019-12-31

    申请号：US16417190

    申请日：2019-05-20

    Applicant: Google LLC

    Inventor： Noam M. Shazeer ,  Jakob D. Uszkoreit ,  Mitchell Thomas Stern

    IPC: G06F15/18 ,  G06K9/62 ,  G06N20/00 ,  G06N7/00

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing parallel generation of output from an autoregressive sequence to sequence model. In one aspect, a blockwise parallel decoding method takes advantage of the fact that some architectures can score sequences in sublinear time. By generating predictions for multiple time steps at once then backing off to a longest prefix validated by the scoring model, the methods can substantially improve the speed of greedy decoding without compromising performance.

    GENERATING NEURAL NETWORK OUTPUTS USING INSERTION COMMANDS
    5.
    发明公开
    GENERATING NEURAL NETWORK OUTPUTS USING INSERTION COMMANDS 审中-公开

    公开(公告)号：US20240028893A1

    公开(公告)日：2024-01-25

    申请号：US18321696

    申请日：2023-05-22

    Applicant: Google LLC

    Inventor： William Chan ,  Mitchell Thomas Stern ,  Nikita Kitaev ,  Kelvin Gu ,  Jakob D. Uszkoreit

    IPC: G06N3/08 ,  G06F40/237 ,  G06N3/04 ,  G06N3/084

    CPC classification number: G06N3/08 ,  G06F40/237 ,  G06N3/04 ,  G06N3/084

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing sequence modeling tasks using insertions. One of the methods includes receiving a system input that includes one or more source elements from a source sequence and zero or more target elements from a target sequence, wherein each source element is selected from a vocabulary of source elements and wherein each target element is selected from a vocabulary of target elements; generating a partial concatenated sequence that includes the one or more source elements from the source sequence and the zero or more target elements from the target sequence, wherein the source and target elements arranged in the partial concatenated sequence according to a combined order; and generating a final concatenated sequence that includes a finalized source sequence and a finalized target sequence, wherein the finalized target sequence includes one or more target elements.

    PARALLEL DECODING USING TRANSFORMER MODELS
    8.
    发明申请
    PARALLEL DECODING USING TRANSFORMER MODELS 审中-公开

    公开(公告)号：US20200082226A1

    公开(公告)日：2020-03-12

    申请号：US16682611

    申请日：2019-11-13

    Applicant: Google LLC

    Inventor： Noam M. Shazeer ,  Jakob D. Uszkoreit ,  Mitchell Thomas Stern

    IPC: G06K9/62 ,  G06N7/00 ,  G06N20/00

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing parallel generation of output from an autoregressive sequence to sequence model. In one aspect, a blockwise parallel decoding method takes advantage of the fact that some architectures can score sequences in sublinear time. By generating predictions for multiple time steps at once then backing off to a longest prefix validated by the scoring model, the methods can substantially improve the speed of greedy decoding without compromising performance.

    Generating neural network outputs using insertion commands
    9.
    发明授权
    Generating neural network outputs using insertion commands 有权

    公开(公告)号：US12086715B2

    公开(公告)日：2024-09-10

    申请号：US18321696

    申请日：2023-05-22

    Applicant: Google LLC

    Inventor： William Chan ,  Mitchell Thomas Stern ,  Nikita Kitaev ,  Kelvin Gu ,  Jakob D. Uszkoreit

    IPC: G06F40/30 ,  G06F40/237 ,  G06N3/04 ,  G06N3/08 ,  G06N3/084

    CPC classification number: G06N3/08 ,  G06F40/237 ,  G06N3/04 ,  G06N3/084

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing sequence modeling tasks using insertions. One of the methods includes receiving a system input that includes one or more source elements from a source sequence and zero or more target elements from a target sequence, wherein each source element is selected from a vocabulary of source elements and wherein each target element is selected from a vocabulary of target elements; generating a partial concatenated sequence that includes the one or more source elements from the source sequence and the zero or more target elements from the target sequence, wherein the source and target elements arranged in the partial concatenated sequence according to a combined order; and generating a final concatenated sequence that includes a finalized source sequence and a finalized target sequence, wherein the finalized target sequence includes one or more target elements.

    Generating neural network outputs using insertion commands
    10.
    发明授权
    Generating neural network outputs using insertion commands 有权

    公开(公告)号：US11657277B2

    公开(公告)日：2023-05-23

    申请号：US16883772

    申请日：2020-05-26

    Applicant: Google LLC

    Inventor： William Chan ,  Mitchell Thomas Stern ,  Nikita Kitaev ,  Kelvin Gu ,  Jakob D. Uszkoreit

    IPC: G10L25/00 ,  G10L21/00 ,  G06F40/30 ,  G06N3/08 ,  G06F40/237 ,  G06N3/04 ,  G06N3/084

    CPC classification number: G06N3/08 ,  G06F40/237 ,  G06N3/04 ,  G06N3/084

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing sequence modeling tasks using insertions. One of the methods includes receiving a system input that includes one or more source elements from a source sequence and zero or more target elements from a target sequence, wherein each source element is selected from a vocabulary of source elements and wherein each target element is selected from a vocabulary of target elements; generating a partial concatenated sequence that includes the one or more source elements from the source sequence and the zero or more target elements from the target sequence, wherein the source and target elements arranged in the partial concatenated sequence according to a combined order; and generating a final concatenated sequence that includes a finalized source sequence and a finalized target sequence, wherein the finalized target sequence includes one or more target elements.

Patent Agency Ranking