-
11.发明授权公开(公告)号：US12211304B2
公开(公告)日：2025-01-28
申请号：US17200448
申请日：2021-03-12
Inventor： Yulin Li , Xiameng Qin , Chengquan Zhang , Junyu Han , Errui Ding , Tian Wu , Haifeng Wang
IPC: G06F16/901 , G06N3/047 , G06N5/04 , G06V10/22 , G06V10/80 , G06V30/262 , G06V30/414 , G06V10/24
Abstract: Embodiments of the present disclosure provide a method and apparatus for performing a structured extraction on a text, a device and a storage medium. The method may include: performing a text detection on an entity text image to obtain a position and content of a text line of the entity text image; extracting multivariate information of the text line based on the position and the content of the text line; performing a feature fusion on the multivariate information of the text line to obtain a multimodal fusion feature of the text line; performing category and relationship reasoning based on the multimodal fusion feature of the text line to obtain a category and a relationship probability matrix of the text line; and constructing structured information of the entity text image based on the category and the relationship probability matrix of the text line.
-
12.发明授权公开(公告)号：US11914964B2
公开(公告)日：2024-02-27
申请号：US17209124
申请日：2021-03-22
Inventor： Shuohuan Wang , Jiaxiang Liu , Xuan Ouyang , Yu Sun , Hua Wu , Haifeng Wang
Abstract: The present application discloses a method and apparatus for training a semantic representation model, a device and a computer storage medium, which relates to the field of natural language processing technologies in artificial intelligence. An implementation includes: acquiring a semantic representation model which has been trained for a first language as a first semantic representation model; taking a bottom layer and a top layer of the first semantic representation model as trained layers, initializing the trained layers, keeping model parameters of other layers unchanged, and training the trained layers using training language materials of a second language until a training ending condition is met; successively bringing the untrained layers into the trained layers from bottom to top, and executing these layers respectively: keeping the model parameters of other layers than the trained layers unchanged, and training the trained layers using the training language materials of the second language until the training ending condition is met respectively; and obtaining a semantic representation model for the second language after all the layers are trained.
-
公开(公告)号：US20230409626A1
公开(公告)日：2023-12-21
申请号：US17754464
申请日：2021-07-20
Inventor： Jizhou Huang , Yibo Sun , Haifeng Wang
IPC: G06F16/387 , G06F40/295
CPC classification number: G06F16/387 , G06F40/295
Abstract: The present disclosure discloses a method and apparatus for acquiring point of interest (POI) state information, and relates to a big data technology in the technical field of artificial intelligence. A specific implementation scheme involves: acquiring a text including POI information within a preset period from the Internet; and recognizing the text by using a pre-trained POI state recognition model, to obtain a two-tuple in the text, the two-tuple including a POI name and POI state information corresponding to the POI name The POI state recognition model acquires a vector representation of each first semantic unit in the text, and acquires a vector representation of each second semantic unit in the text based on semantic dependency information of the text; fuses the vector representation of each first semantic unit and the vector representation of each second semantic unit to obtain a fusion vector representation of each first semantic unit; and predicts labels of the POI name and a POI state based on the fusion vector representation of each first semantic unit. The technical solutions of the present disclosure can save labor costs and improve timeliness and accuracy.
-
公开(公告)号：US11615613B2
公开(公告)日：2023-03-28
申请号：US17210188
申请日：2021-03-23
Inventor： Jizhou Huang , Hao Zhang , Haifeng Wang
Abstract: The present application discloses a method and apparatus for extracting a geographic location point spatial relationship, and relates to the field of big data technologies. A specific implementation solution is as follows: determining geographic location point pairs included in real-scene images by performing signboard recognition on the real-scene images collected by terminal devices; acquiring at least two real-scene images collected by the same terminal device and including the same geographic location point pair; and determining a spatial relationship of the same geographic location point pair by using shooting parameters of the at least two real-scene images. The geographic location point spatial relationship extracted through the present application has higher accuracy and a coverage rate.
-
公开(公告)号：US11574146B2
公开(公告)日：2023-02-07
申请号：US17104165
申请日：2020-11-25
Inventor： Long Li , Haifeng Wang , Weibao Gong
Abstract: A method for updating a parameter of a model, a distributed training system, and an electric device are related to a field of deep learning technologies. The method includes: obtaining a batch training period of batch training data to be trained for a model; increasing priorities of tasks ranked at a bottom in a sequence of gradient communication tasks for parameters of the model when the batch training period is greater than or equal to a preset period threshold; and performing a communication of gradients of the parameters and updating the parameters based on priorities of the gradient communication tasks for the parameters in the model.
-
Patent Agency Ranking
公开(公告)号：US11574135B2
公开(公告)日：2023-02-07
申请号：US16861750
申请日：2020-04-29
Inventor： Haifeng Wang , Hua Wu , Zhongjun He , Hao Xiong
Abstract: The present disclosure provides a method, apparatus, electronic device and readable storage medium for translation and relates to translation technologies. In the embodiments of the present disclosure, the at least one knowledge element is obtained according to associated information of content to be translated, and respective knowledge element in the at least one knowledge element comprise an element of the first language type and an element of the second language type so that the at least one knowledge element can be used to obtain a translation result of the content to be translated. Since the at least one knowledge element obtained in advance is taken as global information of the translation task of this time, it can be ensured that the translation result of the same content to be translated is consistent, thereby improving the quality of the translation result.
-
公开(公告)号：US11553048B2
公开(公告)日：2023-01-10
申请号：US17824318
申请日：2022-05-25
Inventor： Yunpeng Peng , Haifeng Wang
IPC: H04L29/08 , H04L67/306 , H04L67/50 , G06K9/62
Abstract: An object recommendation method, computer device, and medium are provided, relating to the field of artificial intelligence and, particularly, content recommendation. A method includes: obtaining a first user profile of a user, the first user profile being determined based on behavior data of the user over a first historical period of time; using a matching model to determine a recommended object based on the first user profile; recommending the recommended object to the user; obtaining a second user profile of the user, the second user profile being determined based on behavior data of the user over a second historical period of time, and the behavior data of the user over the second historical period of time includes behavior data of the user after the recommended object is recommended to the user; and updating the matching model based on the first user profile, the second user profile, and the recommended object.
-
公开(公告)号：US20220067372A1
公开(公告)日：2022-03-03
申请号：US17210188
申请日：2021-03-23
Inventor： Jizhou Huang , Hao Zhang , Haifeng Wang
Abstract: The present application discloses a method and apparatus for extracting a geographic location point spatial relationship, and relates to the field of big data technologies. A specific implementation solution is as follows: determining geographic location point pairs included in real-scene images by performing signboard recognition on the real-scene images collected by terminal devices; acquiring at least two real-scene images collected by the same terminal device and including the same geographic location point pair; and determining a spatial relationship of the same geographic location point pair by using shooting parameters of the at least two real-scene images. The geographic location point spatial relationship extracted through the present application has higher accuracy and a coverage rate.
-
19.发明申请公开(公告)号：US20220004716A1
公开(公告)日：2022-01-06
申请号：US17209124
申请日：2021-03-22
Inventor： Shuohuan Wang , Jiaxiang Liu , Xuan Ouyang , Yu Sun , Hua Wu , Haifeng Wang
Abstract: The present application discloses a method and apparatus for training a semantic representation model, a device and a computer storage medium, which relates to the field of natural language processing technologies in artificial intelligence. An implementation includes: acquiring a semantic representation model which has been trained for a first language as a first semantic representation model; taking a bottom layer and a top layer of the first semantic representation model as trained layers, initializing the trained layers, keeping model parameters of other layers unchanged, and training the trained layers using training language materials of a second language until a training ending condition is met; successively bringing the untrained layers into the trained layers from bottom to top, and executing these layers respectively: keeping the model parameters of other layers than the trained layers unchanged, and training the trained layers using the training language materials of the second language until the training ending condition is met respectively; and obtaining a semantic representation model for the second language after all the layers are trained.
-
20.发明授权公开(公告)号：US11182648B2
公开(公告)日：2021-11-23
申请号：US16901940
申请日：2020-06-15
Inventor： Hao Xiong , Zhongjun He , Zhi Li , Hua Wu , Haifeng Wang
IPC: G06K9/62 , G06F40/117 , G06N20/00
Abstract: The present disclosure provides an end-to-end model training method and apparatus, which relates to a field of artificial intelligence technologies. The method includes: obtaining training data containing a plurality of training samples, in which the plurality of training samples include an original sequence, a target sequence and a corresponding tag list, the tag list includes importance tags in the target sequence and avoidance tags corresponding to the importance tags, and the avoidance tags are irrelevant tags corresponding to the importance tags; and adopting the training data to train a preset end-to-end model until a value of a preset optimization target function is smaller than a preset threshold.
-
-
-
-
-
-
-
-
-
Search Results
80
records
(took 0.178 seconds)
