公开(公告)号：US20230089268A1

公开(公告)日：2023-03-23

申请号：US18060692

申请日：2022-12-01

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Hao Li ,  Zhenyu Jiao ,  Shuqi Sun ,  Yue Chang ,  Tingting Li

IPC: G06F40/56 ,  G06F40/35 ,  G06F16/332

Abstract: A semantic understanding method, includes: acquiring a query statement and a preceding dialogue; rewriting the query statement based on a preset rule to generate a target query statement if it is recognized that the query statement meets a rule rewriting condition according to the query statement and the preceding dialogue; rewriting the query statement based on a rewriting model to generate the target query statement if it is recognized that the query statement does not meet the rule rewriting condition according to the query statement and the preceding dialogue; and performing intention recognition according to the target query statement to generate an intention recognition result.

公开(公告)号：US20230004721A1

公开(公告)日：2023-01-05

申请号：US17655770

申请日：2022-03-21

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Shuai Zhang ,  Lijie Wang ,  Xinyan Xiao ,  Yue Chang

IPC: G06F40/30 ,  G06F40/211 ,  G06F40/284

Abstract: Disclosed are a method for training a semantic representation model, a device and a storage medium, which relate to the field of computer technologies, and particularly to the field of artificial intelligence, such as a natural language processing technology, a deep learning technology, or the like. The method for training a semantic representation model includes: obtaining an anchor sample based on a sentence, and obtaining a positive sample and a negative sample based on syntactic information of the sentence; processing the anchor sample, the positive sample and the negative sample using the semantic representation model respectively, so as to obtain an anchor-sample semantic representation, a positive-sample semantic representation and a negative-sample semantic representation; constructing a contrast loss function based on the anchor-sample semantic representation, the positive-sample semantic representation, and the negative-sample semantic representation; and training the semantic representation model based on the contrast loss function.

公开(公告)号：US12086555B2

公开(公告)日：2024-09-10

申请号：US17643053

申请日：2021-12-07

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Jianglu Hu ,  Hehan Li ,  Huifeng Sun ,  Shuqi Sun ,  Yue Chang ,  Tingting Li ,  Hua Wu ,  Haifeng Wang

IPC: G06F40/35 ,  G06F16/332

CPC classification number: G06F40/35 ,  G06F16/3329

Abstract: The disclosure provides a method for generating a dialogue. The method includes: obtaining an input sentence; determining a type of a task-based response sentence that is to be generated, by updating a current dialogue state based on the input sentence; generating the task-based response sentence by inputting the input sentence into a task-based dialogue response generator; and determining the task-based response sentence as a target response sentence in response to the type of the task-based response sentence being a designated type.

公开(公告)号：US11989516B2

公开(公告)日：2024-05-21

申请号：US17572068

申请日：2022-01-10

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Lijie Wang ,  Shuai Zhang ,  Xinyan Xiao ,  Yue Chang ,  Tingting Li

IPC: G06F40/289 ,  G06N20/00

CPC classification number: G06F40/289 ,  G06N20/00

Abstract: The present disclosure provides a method and apparatus for acquiring a pre-trained model, an electronic device and a storage medium, and relates to the field of artificial intelligence, such as the natural language processing field, the deep learning field, or the like. The method may include: adding, in a process of training a pre-trained model using training sentences, a learning objective corresponding to syntactic information for a self-attention module in the pre-trained model; and training the pre-trained model according to the defined learning objective. The solution of the present disclosure may improve a performance of the pre-trained model, and reduce consumption of computing resources, or the like.

公开(公告)号：US20230094730A1

公开(公告)日：2023-03-30

申请号：US18061029

申请日：2022-12-02

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Hongyang Zhang ,  Zhenyu Jiao ,  Shuqi Sun ,  Yue Chang ,  Tingting Li

IPC: G06F40/35 ,  G06F40/186 ,  G06F40/279

Abstract: A model training method, a method and an apparatus for human-machine interaction. The method includes: acquiring a sample set corresponding to a template; constructing positive example pairs and negative example pairs for a contrastive learning task based on the sample set; performing contrastive learning training on a pre-training model based on the positive example pairs and the negative example pairs for the contrastive learning task.