利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

310 条记录 (耗时 0.019 秒)

    ML USING N-GRAM INDUCED INPUT REPRESENTATION
    13.
    发明公开
    ML USING N-GRAM INDUCED INPUT REPRESENTATION 审中-公开

    公开(公告)号：US20240086619A1

    公开(公告)日：2024-03-14

    申请号：US18384322

    申请日：2023-10-26

    申请人： Microsoft Technology Licensing, LLC

    发明人： Pengcheng HE ,  Xiaodong Liu ,  Jianfeng Gao ,  Weizhu Chen

    IPC分类号： G06F40/126 ,  G06F40/151 ,  G06N3/08

    CPC分类号： G06F40/126 ,  G06F40/151 ,  G06N3/08

    摘要： Generally discussed herein are devices, systems, and methods for generating an embedding that is both local string dependent and global string dependent. The generated embedding can improve machine learning (ML) model performance. A method can include converting a string of words to a series of tokens, generating a local string-dependent embedding of each token of the series of tokens, generating a global string-dependent embedding of each token of the series of tokens, combining the local string dependent embedding the global string dependent embedding to generate an n-gram induced embedding of each token of the series of tokens, obtaining a masked language model (MLM) previously trained to generate a masked word prediction, and executing the MLM based on the n-gram induced embedding of each token to generate the masked word prediction.

    ML using n-gram induced input representation
    15.
    发明授权
    ML using n-gram induced input representation 有权

    公开(公告)号：US11836438B2

    公开(公告)日：2023-12-05

    申请号：US17229140

    申请日：2021-04-13

    申请人： Microsoft Technology Licensing, LLC

    发明人： Pengcheng He ,  Xiaodong Liu ,  Jianfeng Gao ,  Weizhu Chen

    IPC分类号： G06F40/126 ,  G06N3/08 ,  G06F40/151

    CPC分类号： G06F40/126 ,  G06F40/151 ,  G06N3/08

    摘要： Generally discussed herein are devices, systems, and methods for generating an embedding that is both local string dependent and global string dependent. The generated embedding can improve machine learning (ML) model performance. A method can include converting a string of words to a series of tokens, generating a local string-dependent embedding of each token of the series of tokens, generating a global string-dependent embedding of each token of the series of tokens, combining the local string dependent embedding the global string dependent embedding to generate an n-gram induced embedding of each token of the series of tokens, obtaining a masked language model (MLM) previously trained to generate a masked word prediction, and executing the MLM based on the n-based induced embedding of each token to generate the masked word prediction.

    SYSTEMS AND METHODS OF A SCRIPT GENERATION ENGINE
    16.
    发明公开
    SYSTEMS AND METHODS OF A SCRIPT GENERATION ENGINE 审中-公开

    公开(公告)号：US20230367954A1

    公开(公告)日：2023-11-16

    申请号：US18347949

    申请日：2023-07-06

    申请人： Open Text SA ULC

    发明人： Gregory R. Petti

    IPC分类号： G06F40/143 ,  G06F8/40 ,  G06F40/151

    CPC分类号： G06F40/143 ,  G06F8/40 ,  G06F40/151 ,  G06F9/44

    摘要： A template built by a user may be converted by a Server Script Generation Engine (SSGE) into script code. In converting, the SSGE may load and parse a framework file containing static script syntax to locate insertion points, each associated with an iteration number, and may iteratively parse the template, utilizing the iteration number to resolve, in order, tags and sub-tags contained in the template. If a tag is set to respond to the iteration number, a function of the tag is invoked to process any related sub-tags and return a script associated therewith at the appropriate insertion point. The framework file (with the appropriate script code inserted) is compiled and stored in a compiled script object which can be run multiple times to perform all of the output functions expected by the user in lieu of the need to reconvert the template.

    MULTILINGUAL SUMMARIZATION OF EPISODES USING LONGFORMERS
    17.
    发明公开
    MULTILINGUAL SUMMARIZATION OF EPISODES USING LONGFORMERS 审中-公开

    公开(公告)号：US20230359837A1

    公开(公告)日：2023-11-09

    申请号：US17735384

    申请日：2022-05-03

    申请人： Spotify AB

    发明人： Edgar Tanaka ,  Ann Clifton

    IPC分类号： G06F40/58 ,  G06F40/263 ,  G06F40/284 ,  G06F40/151 ,  G06F40/166 ,  G06F40/197 ,  G06N20/10

    CPC分类号： G06F40/58 ,  G06F40/263 ,  G06F40/284 ,  G06F40/151 ,  G06F40/166 ,  G06F40/197 ,  G06N20/10

    摘要： A full attention mechanism of a multilingual transformer model is converted into a Longformer attention mechanism to generate a Longformer multilingual transformer model. The Longformer multilingual transformer model is finetuned to perform a summarization task based on episode-description:episode-transcript pairs, thereby generating a finetuned Longformer multilingual transformer model. The Longformer multilingual transformer model also can further be finetuned to perform a summarization task based on article-summary:full-original-article pairs. A summary of a query episode transcript can be generated using the single-finetuned Longformer multilingual transformer model and/or the double-finetuned Longformer multilingual transformer model. The multilingual transformer-based model enables systems, methods and computer products to be capable of generating multilingual abstractive summaries.

    Systems and methods of a script generation engine
    19.
    发明授权
    Systems and methods of a script generation engine 有权

    公开(公告)号：US11741293B2

    公开(公告)日：2023-08-29

    申请号：US17558070

    申请日：2021-12-21

    申请人： Open Text SA ULC

    发明人： Gregory R. Petti

    IPC分类号： G06F40/00 ,  G06F40/143 ,  G06F8/40 ,  G06F40/151 ,  G06F9/44 ,  G06F8/41 ,  G06F9/455 ,  G06F40/154

    CPC分类号： G06F40/143 ,  G06F8/40 ,  G06F40/151 ,  G06F8/41 ,  G06F9/44 ,  G06F9/45512 ,  G06F40/154

    摘要： A template built by a user may be converted by a Server Script Generation Engine (SSGE) into script code. In converting, the SSGE may load and parse a framework file containing static script syntax to locate insertion points, each associated with an iteration number, and may iteratively parse the template, utilizing the iteration number to resolve, in order, tags and sub-tags contained in the template. If a tag is set to respond to the iteration number, a function of the tag is invoked to process any related sub-tags and return a script associated therewith at the appropriate insertion point. The framework file (with the appropriate script code inserted) is compiled and stored in a compiled script object which can be run multiple times to perform all of the output functions expected by the user in lieu of the need to reconvert the template.