-
公开(公告)号：US20240411790A1
公开(公告)日：2024-12-12
申请号：US18747415
申请日：2024-06-18
Inventor： Zhuangzhuang Cui , Bo Fu
IPC: G06F16/332 , G06F16/33 , G06F16/35 , G06F40/186 , G06F40/30
Abstract: The present disclosure provides an answer information generation method and apparatus based on a large language model, and a device, and relates to the technical field of artificial intelligence, and in particular, to the fields of document retrieval, natural language processing, and large language models. An implementation solution includes: obtaining, in response to receiving a question text from a user, a semantic vector of the question text and event information related to a specific field; obtaining a plurality of candidate documents from a document library of the specific field based on at least two of the semantic vector of the question text, the at least one piece of argument information and the event category; determining quality evaluation information for a candidate document in the plurality of candidate documents based on the event category; and determining at least one target document from the plurality of candidate documents based on the quality evaluation information of the candidate document and a correlation between the candidate document and the question text, to obtain, based on the at least one target document, answer information used to answer the question text.
-
公开(公告)号：US20240403344A1
公开(公告)日：2024-12-05
申请号：US18803085
申请日：2024-08-13
Inventor： Wanpeng NIU , Junwei XING , Sai GAO , Haonan FANG , Hui LI , Bingfei ZHANG
Abstract: There is provided a code retrieval method and apparatus based on a large language model, an electronic device and a readable storage medium, which relates to the field of artificial intelligence technologies, such as large language model technologies, big data technologies, cloud service technologies, or the like. The method for code retrieval based on a large language model includes: acquiring a code retrieval query to obtain a retrieval vector of the code retrieval query; acquiring a first index of a target code library, the first index including a plurality of code blocks and a plurality of code block vectors; acquiring a target code block according to the retrieval vector and the first index; acquiring a second index of the target code library, the second index being a code architecture knowledge graph; acquiring a target code file corresponding to the target code block according to a source code file corresponding to the target code block and the second index; and acquiring a retrieval result according to the target code block and the target code file.
-
公开(公告)号：US12158801B2
公开(公告)日：2024-12-03
申请号：US18157429
申请日：2023-01-20
Inventor： Zhigang Zeng , Zhenyuan Sun , Bingqing Shao , Pengfei Yan , Shiyong Li , Yanpeng Wang
IPC: G06F11/07
Abstract: A method of responding to an operation, an electronic device and a storage medium are provided, which relate to a field of cloud computing, and in particular to a field of cluster technology. The specific implementation solution includes: performing, in response to determining that a target operation performed by a target client on a shared resource has timed out, a fault detection on the target client to obtain a fault detection result; and implementing, in response to determining that the fault detection result represents that the target client has a fault, an update operation to obtain a target authority identifier, so that the target client is prevent from continuing to perform the target operation by using the target authority identifier.
-
公开(公告)号：US12147448B2
公开(公告)日：2024-11-19
申请号：US18088712
申请日：2022-12-26
Inventor： Zhengli Yi
Abstract: The present disclosure provides a data reading method, including: in response to receiving a read response request generated by a replication group for an application and sent by a storage terminal, setting, in a dedicated mapping table corresponding to the application, a commit index corresponding to the replication group as a commit index carried by the read response request; searching for a target replication group corresponding to a first read request generated by a target application based on the first read request; determining, in a dedicated mapping table corresponding to the target application, a commit index corresponding to the target replication group as a target commit index; sending a second read request carrying the target commit index to the storage terminal; and obtaining the data read by the storage terminal.
-
公开(公告)号：US20240378077A1
公开(公告)日：2024-11-14
申请号：US18782617
申请日：2024-07-24
Inventor： Guoxia WANG , Jinle ZENG , Xiyuan XIAO , Jiabin YANG , Dianhai YU , Haifeng WANG
Abstract: A method of executing a task for a large language model, a device, and a storage medium are provided, which relate to a field of artificial intelligence technology, and in particular to fields of deep learning, large language model, natural language processing and computer vision technologies. The method includes: determining, by using a determination unit, a target attention task from a plurality of attention tasks to be processed, based on a sparse representation corresponding to a feature to be processed, where the target attention task is a task corresponding to a non-fully masked region of the feature, the sparse representation represents a mask position of the feature, and the mask position represents mask endpoint positions in at least two non-intersecting intervals in a mask matrix corresponding to the feature; and executing the target attention task by using a computing unit, so as to obtain an attention feature.
-
Patent Agency Ranking
公开(公告)号：US12140441B2
公开(公告)日：2024-11-12
申请号：US17531529
申请日：2021-11-19
Inventor： Weijia Zhang , Hao Liu , Dejing Dou , Hui Xiong
IPC: G01C21/34 , G06N20/00 , G06Q30/0601 , H04L67/12
Abstract: A method for recommending a station for a vehicle, a device, and a storage medium are provided. The method comprises: receiving, by a server, an access request from a vehicle; obtaining, based on the access request, a plurality of observation values from a plurality of stations associated with the vehicle, respectively, each observation value is based on a corresponding pre-trained recommendation model, each observation value includes factors associated with access of the vehicle to the station corresponding to the observation value; determining, an action value for the station based on the observation value and the pre-trained recommendation model for the station, the action value for the station indicates a matching degree between the access request and the station; determining a recommended station among the plurality of stations based on the action values of the plurality of stations; and sending to the vehicle an instruction of driving to the recommended station.
-
337.发明授权公开(公告)号：US12125131B2
公开(公告)日：2024-10-22
申请号：US18075346
申请日：2022-12-05
Inventor： Zhe Peng , Yuqiang Liu , Fanyu Geng
CPC classification number: G06T13/40 , G06T7/20 , G10L15/02 , G10L15/063 , G10L15/16 , G10L25/57 , G06T2207/20081 , G06T2207/20084 , G06T2207/30201
Abstract: A method of generating a 3D video, a method of training a neural network model, an electronic device, and a storage medium, which relate to a field of image processing, and in particular to technical fields of computer vision, augmented/virtual reality and deep learning. The method includes: determining, based on an input speech feature, a principal component analysis (PCA) coefficient by using a first network, wherein the PCA coefficient is used to generate the 3D video; correcting the PCA coefficient by using a second network; generating a lip movement information based on the corrected PCA coefficient and a PCA parameter for a neural network model, wherein the neural network model includes the first network and the second network; and applying the lip movement information to a pre-constructed 3D basic avatar model to obtain a 3D video with a lip movement effect.
-
公开(公告)号：US12118992B2
公开(公告)日：2024-10-15
申请号：US17622402
申请日：2021-06-02
Inventor： Jizhou Huang , Shiqiang Ding
IPC: G10L15/22 , G06F16/9032 , G10L15/18 , G10L15/30
CPC classification number: G10L15/22 , G06F16/90332 , G10L15/1815 , G10L15/30 , G10L2015/223
Abstract: Technical solutions relate to the fields of artificial intelligence technologies and voice technologies. A technical solution includes: performing voice recognition and demand analysis on a voice instruction input by a user; in response to an unknown demand obtained by the demand analysis, acquiring information of a query entity and query content using a result of the demand analysis, and acquiring reply information corresponding to the query content by communication with the query entity; and returning a first voice response to the user using the reply information.
-
339.发明授权公开(公告)号：US12118319B2
公开(公告)日：2024-10-15
申请号：US17655772
申请日：2022-03-21
Inventor： Jun Xu , Zeming Liu , Zeyang Lei , Zhengyu Niu , Hua Wu , Haifeng Wang
Abstract: The present disclosure provides a dialog method and system, an electronic device and a storage medium, and relates to the field of artificial intelligence (AI) technologies such as deep learning and natural language processing. A specific implementation scheme involves: rewriting a corresponding dialog state based on received dialog information of a user; determining to-be-used dialog action information based on the dialog information of the user and the dialog state; and generating a reply statement based on the dialog information of the user and the dialog action information. According to the present disclosure, the to-be-used dialog action information can be determined based on the dialog information of the user and the dialog state; and then the reply statement is generated based on the dialog action information, thereby providing an efficient dialog scheme.
-
340.发明公开公开(公告)号：US20240338564A1
公开(公告)日：2024-10-10
申请号：US18744501
申请日：2024-06-14
Inventor： Zhifan FENG , Hua WU , Qiaoqiao SHE , Tian WU
IPC: G06N3/08
CPC classification number: G06N3/08
Abstract: A large model optimization training method in the artificial intelligence fields, such as large models, deep learning, natural language processing, may include: taking, as candidate queries, queries collected from a predetermined data source and capable of serving as input to a large model in response to determining that an optimization triggering condition is met; screening out target queries from the candidate queries, the target queries being queries which cannot be correctly processed by the large model; and constructing respectively corresponding training samples according to the target queries, the training samples being used for carrying out optimization training on the large model.
-
-
-
-
-
-
-
-
-
Search Results
1522
records
(took 0.147 seconds)
