Method and apparatus for generating dialogue models
A technology of models and training samples, which is applied in the computer field, can solve problems such as lack of information, inability of users to answer effectively, single reply content, etc., and achieve the effect of increasing diversity
Active Publication Date: 2021-08-10
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The reply content generated by this dialogue model is single and lacks information, and cannot accurately and effectively answer the above text sent by the user.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0024] The present disclosure will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain related inventions, rather than to limit the invention. It should also be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.
[0025] It should be noted that, in the case of no conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and embodiments.
[0026] figure 1 An exemplary system architecture 100 is shown in which the method for generating a dialog model or the device for generating a dialog model of the embodiments of the present disclosure can be applied.
[0027] Such as figure 1 As shown, the system arch...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
Embodiments of the present disclosure disclose methods and devices for generating dialogue models. A specific implementation of the method includes: obtaining a corpus sample set, wherein the corpus sample includes input information and target reply information; classifying the corpus samples in the corpus sample set, setting discrete latent variables for each corpus sample according to the classification results, and generating training A sample set, wherein the training samples include input information, target reply information and discrete hidden variables; use the training sample set to train the preset neural network to obtain a dialogue model, wherein the dialogue model is used to represent the input input information and the output target reply information Correspondence between. In this embodiment, the "one-to-many" relationship between dialogue texts can be effectively modeled by introducing discrete latent variables, which improves the diversity of dialogue model replies, thereby enabling effective answers to the above given by users.
Description
technical field [0001] The embodiments of the present disclosure relate to the field of computer technology, and in particular to a method and device for generating a dialogue model. Background technique [0002] Dialogue models can generally be applied to intelligent dialogue systems, such as chat systems that can be used for user communication, or navigation robots, etc. Specifically, for the above text sent by the user (which can be in the form of text or voice), the dialogue model can perform internal calculations on the text given by the user, determine the reply to the above text, and then use the reply as the model output feedback to the user. Generally, the dialogue model may use a sequence to sequence (sequence to sequence, seq2seq for short) model to realize the above dialogue function (ie, short text reply function). [0003] In related technologies, dialogue models are often based on large-scale pre-trained language models, such as BERT. The reply content gene...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/332G06F40/295G06F40/186G06N3/04G06N3/08G10L15/26
CPCG06F16/3329G06F16/3326G06N3/08G10L15/26G06N3/044G06N3/045G06F40/35G06F40/30
Inventor 鲍思琪何煌陈俊锟王凡吴华何径舟
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD