利索-全球专利检索

  1. Login
  2. Sign Up

Search Results

283 records (took 0.022 seconds)

    System and Method for Measuring and Improving Literacy by Deep Learning and Large Language Model Assisted Reading
    1.
    发明公开
    System and Method for Measuring and Improving Literacy by Deep Learning and Large Language Model Assisted Reading 审中-公开

    公开(公告)号：US20240331558A1

    公开(公告)日：2024-10-03

    申请号：US18611932

    申请日：2024-03-21

    Applicant: Ruhul Quddus

    Inventor： Ruhul Quddus

    IPC: G09B5/06 ,  G06F16/955 ,  G06F40/134 ,  H04L51/02

    CPC classification number: G09B5/06 ,  G06F16/9558 ,  G06F40/134 ,  H04L51/02

    Abstract: The method and system described herein provide a novel safe approach to enhancing and measuring the understanding of technical, science or other documents. By identifying difficult concepts and providing additional safety verified resources to aid comprehension, the system enables readers to better understand complex technical documents. The system may be implemented as a web-based application or as a plug-in for existing reading applications and may be customized for specific domains. The system works on existing documents. Content owners do not have to recreate content to fit a new format. The system and method have broad applicability beyond making it easier for students to understand articles. The system may provide feedback to content authors, enable teachers to identify students' effort in reading a document. It can enable businesses to better engage with their customers while reading their existing marketing and technical contents online.

    Link-based audio recording, collection, collaboration, embedding and delivery system
    3.
    发明授权
    Link-based audio recording, collection, collaboration, embedding and delivery system 有权

    公开(公告)号：US12094446B2

    公开(公告)日：2024-09-17

    申请号：US18228901

    申请日：2023-08-01

    Applicant: NAMECOACH, INC.

    Inventor： Praveen Shanbhag

    IPC: G10L15/22 ,  G06F40/134 ,  G06F40/58 ,  G06Q10/10 ,  G10L13/02

    CPC classification number: G10L13/02 ,  G06F40/134 ,  G06F40/58 ,  G06Q10/10

    Abstract: A machine has a processor and a memory connected to the processor. The memory stores instructions executed by the processor to supply a name page in response to a request from an administrator machine. Name page updates are received from the administrator machine. The name page updates include participants and associated network contact information for the participants. A code is utilized to form a link to the name page. Prompts for textual name information and audio name information are supplied to a client machine that activates the link to the name page. Textual name information and audio name information are received from the client machine. The textual name information and audio name information are stored in association with the name page. Navigation tools are supplied to facilitate access to the textual name information and audio name information.

    CONFIDENCE ENHANCEMENT FOR RESPONSES BY DOCUMENT-BASED LARGE LANGUAGE MODELS
    4.
    发明公开
    CONFIDENCE ENHANCEMENT FOR RESPONSES BY DOCUMENT-BASED LARGE LANGUAGE MODELS 审中-公开

    公开(公告)号：US20240296279A1

    公开(公告)日：2024-09-05

    申请号：US18311973

    申请日：2023-05-04

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Matthew Jonathan GARDNER

    IPC: G06F40/169 ,  G06F40/134 ,  G06F40/279

    CPC classification number: G06F40/169 ,  G06F40/279 ,  G06N20/00 ,  G06F40/134

    Abstract: Systems and methods are provided for implementing confidence enhancement for responses by document-based large language models (“LLMs”) or other AI/ML systems. A first prompt is generated based on data items that are previously received or accessed. The first prompt is used by a first LLM or AI/ML system to extract requested information from the data items. One or more citations are generated and presented within a structured object together with a representation of the extracted information, in some cases, as output from a second LLM or AI/ML system. In some cases, the citations and/or the representation may be verified by a third LLM or AI/ML system, and reliability indicators may be generated for the citations and/or the representation based on determined accuracy of the citations and/or the representation. In this manner, the common issue of hallucinations may be mitigated.

    CUSTOM FORMULAS IN OFFICE CLOUD PLATFORM ADD-IN
    7.
    发明公开
    CUSTOM FORMULAS IN OFFICE CLOUD PLATFORM ADD-IN 审中-公开

    公开(公告)号：US20240256765A1

    公开(公告)日：2024-08-01

    申请号：US18160781

    申请日：2023-01-27

    Applicant: SAP SE

    Inventor： Stephane Coze ,  Dietrich Eisenhardt ,  Bing Sun ,  Laurent Le Floch

    IPC: G06F40/18 ,  G06F40/134 ,  H04L67/02

    CPC classification number: G06F40/18 ,  G06F40/134 ,  H04L67/02

    Abstract: The present disclosure involves systems, software, and computer implemented methods for enabling custom formulas in a cloud platform office add-in. An example method includes automatically loading a public landing page of a cloud platform add-in of a spreadsheet application. The public landing page includes a declaration and a first implementation of at least one custom function. Login credentials for a customer tenant of the cloud platform are obtained. A redirect is performed to a private web application for the customer tenant. The private web application includes, for each custom function, a second implementation that is configured to retrieve customer-specific cloud platform data of the customer tenant. Each first implementation of a custom function is replaced with the second implementation of the custom function. In response to a user request in the spreadsheet application to use a first custom function, the second implementation of the custom function is invoked.

Patent Agency Ranking