Machine translation style migration performance improvement method based on iterative knowledge migration

A machine translation and style technology, applied in the field of machine translation, can solve the problems of restricting the development of stylized machine translation, exacerbating the transmission and accumulation of translation errors, and slowing down the decoding speed, so as to reduce the cost of manual annotation and translation and improve translation Efficiency, performance-enhancing effects

Pending Publication Date: 2021-11-02
GLOBAL TONE COMM TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Both types of methods can improve translation quality, but the first type of method will aggravate the transfer and accumulation of translation errors between models due to two-step decoding, and will slow down the decoding speed; the second type of data enhancement method will The accuracy of style transfer in translation results will be reduced due to the noise in the pseudo-parallel data
These problems greatly limit the development of stylized machine translation
[0007] The difficulty of solving the above problems and defects is: the model training of machine translation needs to be built on large-scale parallel data. In the case of scarce corpus, it is difficult to learn the conversion of language and style at the same time, so there is an urgent need for a method that can make full use of The information of the existing corpus and the ability to use the data to correct errors can improve the quality of the corpus, thereby improving the efficiency and accuracy of stylized machine translation

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
  • Machine translation style migration performance improvement method based on iterative knowledge migration
  • Machine translation style migration performance improvement method based on iterative knowledge migration
  • Machine translation style migration performance improvement method based on iterative knowledge migration

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] A method for improving the performance of machine translation style transfer based on iterative knowledge transfer specifically includes the following steps:

[0073] 1) In the field of general machine translation and specific text style transfer with training data, pre-train the machine translation model and text style transfer model.

[0074] 2) Use the text style transfer model as the teacher model to decode the sentences in the source style and generate the text in the target style.

[0075] 3) The source language sentence and the target style sentence decoded in step 2) can construct a translation pseudo-parallel sentence pair from the source style to the target style for the training of the stylized translation model

[0076] 4) Use the translation style transfer model as the teacher model to decode the sentences in the source language, and translate the text in the target language and target style.

[0077] 5) The source style target language sentence and the ta...

Embodiment 2

[0098] The method for improving the performance of machine translation style transfer according to the second embodiment of the present invention includes the following steps:

[0099] 1) In the field of general machine translation and specific text style transfer with training data, pre-train the machine translation model and text style transfer model.

[0100] Here, the translation model and the text style transfer model can be a sequence-to-sequence structure based on a recurrent neural network, or a Transformer-based self-attention model. This process trains the machine translation model in a semi-supervised manner, using a large amount of monolingual data on the Internet to make up for parallel corpus and alleviate the problem of insufficient parallel corpus for translation. This process uses transfer learning to train the text style transfer model. By fine-tuning the text style transfer data on the pre-trained language model, the knowledge of the pre-trained model is tra...

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 belongs to the technical field of machine translation, and discloses a machine translation style migration performance improvement method based on iterative knowledge migration. The machine translation style migration performance improvement method based on iterative knowledge migration comprises the following steps: pre-training a translation model and a text style migration model, guiding the translation model by the text style migration model, constructing a pseudo-parallel sentence pair and performing data tuning, guiding the text style migration model by the translation model, and iteratively improving the translation style migration performance. According to the method, the problem of less training data in machine translation style migration is relieved. According to the data tuning model, grammar error correction is carried out by fully utilizing the original text and the text after style migration, so that the pseudo-parallel data can be smoother, and the quality of the pseudo-parallel data is effectively improved. The performance of the translation model and the text style migration model is improved.

Description

technical field [0001] The invention belongs to the technical field of machine translation, and in particular relates to a method for improving the performance of machine translation style transfer based on iterative knowledge transfer. Background technique [0002] At present: machine translation refers to the process of translating source language sentences into semantically equivalent target language sentences through computers, and is an important research direction in the field of natural language processing. Machine translation can be mainly divided into three methods: rule-based machine translation, statistics-based machine translation and neural network-based machine translation. Initially, the rule-based method was the mainstream of machine translation research. This method has a good translation effect on sentences with standardized grammatical structures, but it also has the disadvantages of complex rule writing and difficulty in dealing with non-standard language...

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): G06F40/253G06F40/42G06F40/58
CPCG06F40/253G06F40/42G06F40/58
Inventor 李欣杰卢恩全贝超
Owner GLOBAL TONE COMM TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap