公开(公告)号：US20230282016A1

公开(公告)日：2023-09-07

申请号：US17898678

申请日：2022-08-30

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD

Inventor： Huihui HE ,  Jiayang WANG ,  Yubo XIANG

IPC: G06V30/19 ,  G06N3/08 ,  G06T9/00

CPC classification number: G06V30/19187 ,  G06N3/08 ,  G06T9/002 ,  G06V30/19147 ,  G06V30/1916

Abstract: Provided are method for recognizing a receipt, an electronic device and a storage medium, which relate to the fields of deep learning and pattern recognition. The method may include: a target receipt to be recognized is acquired; two-dimensional position information of multiple text blocks on the target receipt respectively is encoded, to obtain multiple encoding results; graph convolution is performed on the multiple encoding results respectively, to obtain multiple convolution results; and each of the multiple convolution results is recognized based on a first conditional random field model, to obtain a first prediction result at text block-level of the target receipt, wherein the first conditional random field model and a second conditional random field model are co-trained, so as to obtain a second prediction result at token-level of the target receipt.

公开(公告)号：US20230142217A1

公开(公告)日：2023-05-11

申请号：US17896690

申请日：2022-08-26

Applicant: Beijing Baidu Netcom Science Technology Co., Ltd.

Inventor： Huihui HE ,  Leyi WANG ,  Duohao QIN ,  Minghao LIU

IPC: G06F40/47 ,  G06F40/166 ,  G06F40/30 ,  G06F40/295 ,  G06F40/151

CPC classification number: G06F40/47 ,  G06F40/166 ,  G06F40/30 ,  G06F40/295 ,  G06F40/151

Abstract: The present disclosure provides a model training method and apparatus, an electronic device, and a storage medium, and relates to the field of artificial intelligence, in particular, to the field of natural language processing and deep learning. A specific implementation solution includes: constructing initial training corpora; performing data enhancement on the initial training corpora based on an algorithm contained in a target algorithm set to obtain target training corpora, wherein the target algorithm set is determined from multiple algorithm sets, and different algorithm sets are used for performing data enhancement on corpora with different granularity in the initial training corpora; and performing training on a language model based on the target training corpora to obtain a sequence labeling model, herein the language model is pre-trained based on text corpora.

公开(公告)号：US20220383190A1

公开(公告)日：2022-12-01

申请号：US17619533

申请日：2021-05-17

Applicant: Beijing Baidu Netcom Science Technology Co., Ltd.

Inventor： Huihui HE ,  Leyi WANG ,  Minghao LIU ,  Jiangliang GUO

IPC: G06N20/00 ,  G06F40/30

Abstract: The present disclosure provides a method of training a classification model, which relates to an active learning, neural network and natural language processing technology. A specific implementation scheme includes: selecting, from an original sample set, a plurality of original samples with a class prediction result meeting a preset condition as to-be-labeled samples according to a class prediction result for a plurality of original samples in the original sample set; labeling the to-be-labeled sample as belonging to a class by using the second classification model, so as to obtain a first labeled sample set; and training the first classification model by using the first labeled sample set. The present disclosure further provides a method of classifying a sample, an electronic device, and a storage medium.

公开(公告)号：US20230281380A1

公开(公告)日：2023-09-07

申请号：US18173651

申请日：2023-02-23

Applicant: Beijing Baidu Netcom Science Technology Co., Ltd.

Inventor： Yubo XIANG ,  Jiayang WANG ,  Huihui HE ,  Junyu SHEN ,  Cuicong SU ,  Hongguang ZHANG

IPC: G06F40/166 ,  G06V30/412 ,  G06F40/284 ,  G06F40/30

CPC classification number: G06F40/166 ,  G06V30/412 ,  G06F40/284 ,  G06F40/30

Abstract: A method of processing a text, an electronic device, and a storage medium, which relates to a field of an image processing technology, in particular to a field of computer vision technology. The method includes: determining similarities between a plurality of fields contained in a text image to be processed and a plurality of predetermined field names; determining, from the plurality of fields, a field having a similarity greater than a similarity threshold as a target field name; determining a target field value corresponding to the target field name from M remaining field among the plurality of fields other than the target field name, where M≥1; and outputting a correspondence between the target field name and the target field value.