Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Negative sample determination method, related equipment and readable storage medium

A determination method and negative sample technology, applied in the field of artificial intelligence, can solve the problems of inability to improve the training effect and poor confusion of negative samples, and achieve the effect of strong confusion, good effect, and improved training effect

Pending Publication Date: 2021-12-28
科大讯飞(北京)有限公司 +2
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the negative sample of the sample problem obtained based on the above method is less confusing and cannot improve the training effect

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Negative sample determination method, related equipment and readable storage medium
  • Negative sample determination method, related equipment and readable storage medium
  • Negative sample determination method, related equipment and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0176] As an implementable manner, the computing unit includes:

[0177] A sample question encoding unit, configured to input the sample question into a question encoder of a semantic similarity calculation model, and the question encoder encodes the sample question to obtain a feature vector of the sample question;

[0178] A text encoding unit, configured to input each text in the text library into the text encoder of the semantic similarity calculation model, and the text encoder encodes each text in the text library to obtain the text library The feature vector of each text in ;

[0179] The semantic similarity calculation unit is used to calculate the feature vector of each text in the text library by using the calculation module of the semantic similarity calculation model, and the similarity with the feature vector of the sample question is used as each text in the text library. The semantic similarity of the text to the sample question in question.

[0180] As an imp...

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

No PUM Login to View More

Abstract

The invention discloses a negative sample determination method, related equipment and a readable storage medium, and can determine a text which has high semantic similarity with a sample question but does not contain an answer to the sample question from a text library as a negative sample of the sample question based on the semantic similarity between each text in the text library and the sample question. The negative samples determined based on the method have high semantic similarity with the sample questions, but do not contain answers to the sample questions, so that the confusability on the sample questions is strong. Furthermore, the negative sample determined based on the method is used for training a retrieval model and a question and answer model in the intelligent question and answer system, and the training effect can be improved. Moreover, when the negative sample determined based on the method is used for testing the robustness of the retrieval model and the question-answering model in the intelligent question-answering system, the effect is better compared with the confrontation sample determined by adopting the prior art.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and more specifically, relates to a negative sample determination method, related equipment, and a readable storage medium. Background technique [0002] With the development of artificial intelligence technology, intelligent question answering systems (such as open domain question answering systems, vertical domain question answering systems, etc.) have been widely used. At present, the commonly used intelligent question answering system includes a retrieval model and a question answering model. When a user searches for a question, the retrieval model will automatically retrieve the text related to the question from the text library, and the question answering model will determine the output based on each text detected by the retrieval model. Answer and return the answer to the user. [0003] When using sample questions to train the retrieval model and question answeri...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/33G06F16/332G06K9/62G06N3/02G06N3/08
CPCG06F16/3329G06F16/3331G06N3/02G06N3/08G06F18/214
Inventor 崔一鸣邵楠王士进刘挺
Owner 科大讯飞(北京)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com