Systems and methods for factual extraction from language model

    公开(公告)号:US12112131B2

    公开(公告)日:2024-10-08

    申请号:US17588043

    申请日:2022-01-28

    申请人: Salesforce, Inc.

    摘要: Embodiments described herein provide a system and method for extracting factual information. The system transforms a query into a natural language prompt in a format of a query subject and a queried relation. The system encodes, via an embedding layer of a pre-trained language model, the natural language prompt into a first embedding. The system encodes, via the adapter model, the first embedding into a second embedding based on a probability that the second embedding returns the factual information when the second embedding is fed the first attention layer of the pre-trained language model. The system decodes, by the first attention layer of the pre-trained language mode, the second embedding into a response to the query. The system extracts the factual information from the decoded response to the query.

    COMMUNICATION OF TEXT DATA THROUGH A VOICE CALL

    公开(公告)号:US20240284148A1

    公开(公告)日:2024-08-22

    申请号:US18170943

    申请日:2023-02-17

    摘要: This document describes techniques, apparatuses, and systems for communication of text data through a voice call. Provided is a voice call between a first user equipment and a second user equipment, during which packets are communicated through a voice channel. Information is requested from the first user equipment. Text conveying the requested information is received at the first user equipment. The text is encoded into data indicative of the text and organized into packets that are marked to indicate the text data within. The packets are transmitted to the second user equipment, where the packets are determined to include text data, and the data indicative of the text is decoded using a decoder associated with text data. In doing so, text data can be communicated through the voice call.

    MULTI-LANGUAGE DOCUMENT FIELD EXTRACTION
    7.
    发明公开

    公开(公告)号:US20240273290A1

    公开(公告)日:2024-08-15

    申请号:US18168450

    申请日:2023-02-13

    申请人: SAP SE

    摘要: A method for multi-language document field extraction may include determining, based on a received document including a plurality of key fields and a plurality of value fields, a plurality of key-value pairs. The method also includes determining whether an encoding of a key field is within a threshold distance from a predetermined encoding of a predefined key field associated with a predefined field type. The method further includes assigning, based on determining the encoding of the key field is within the threshold distance, the predefined field type to the corresponding key-value pair. The method also includes performing a document processing operation based on each key-value pair and the predefined field type assigned to each key-value pair. Related systems and methods are provided.

    REPLY CONTENT PROCESSING METHOD AND INTERACTION METHOD FOR INTERACTIVE CONTENT OF MEDIA CONTENT

    公开(公告)号:US20240265198A1

    公开(公告)日:2024-08-08

    申请号:US18597135

    申请日:2024-03-06

    发明人: Xiaoshuai CHEN

    摘要: A reply content processing method including obtaining to-be-replied interactive content for media content, performing encoding processing on description content of the media content and the to-be-replied interactive content to obtain a vectorized representation of each word in the description content and the to-be-replied interactive content, and performing style recognition based on the vectorized representation of each word in the description content and the to-be-replied interactive content to obtain a first style category set, performing style recognition based on release party information of the to-be-replied interactive content to obtain a second style category set, determining a third style category set to which the to-be-replied interactive content belongs, determining a style category vector corresponding to each style category in the third style category set, and performing reply word prediction based on the description content, the to-be-replied interactive content, and the style category vector to generate reply content corresponding to each style category.

    ENCODING DOCUMENTS FOR PERIPHERAL PRIVACY
    10.
    发明公开

    公开(公告)号:US20240211678A1

    公开(公告)日:2024-06-27

    申请号:US18145412

    申请日:2022-12-22

    IPC分类号: G06F40/126 G06F3/01 G06F21/84

    摘要: A computing system for generating and displaying an encoded document for peripheral privacy is provided. The computing system includes a processor executing a program using portions of memory to determine an initial document, generate an encoded document based on the initial document by modifying letters in a text portion of the initial document, and output the encoded document. In another example, a computing system for displaying an encoded document for peripheral privacy is provided. The computing system includes a processor executing a program using portions of memory to receive an initial document and an encoded document, determine a gaze location of a user using a camera, generate a mask based on the determined gaze location, and display an alpha-blended document by alpha-blending the initial document and the encoded document using the mask.