Model training and data processing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a model training and data processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining training data which comprises a first original text, a first translation and a first post-translation reference translation; training an edit-after-translation model by using the training data; using a pre-trained language model as an encoder of the edit-after-translation model, wherein the initial parameters of the encoder are parameters of the pre-trained language model; enabling the training data to enter a decoder of the edit-after-translation model through the encoder; and adjusting the parameters of the edit-after-translation model according to the output of the decoder. The semantic knowledge between the original text and the translated text corresponding to the original text is learned by using a large amount of pre-training data, and the semantic knowledge is migrated to the training process of the editing-after-translation model, so that the editing-after-translation model has higher robustness, and meanwhile, the problem that the acquisition cost of triples such as training data is relatively high is solved.

Description

technical field [0001] The present disclosure relates to the field of computer technology, and in particular to a model training and data processing method, device, electronic equipment, and storage medium. Background technique [0002] Machine translation refers to the technology of using computer programs to translate sentences from one natural language (source language) to another natural language (target language). The currently more commonly used neural network architecture Transformer is an attention-based encoder-decoder model. The main idea is to encode the sentence to be translated (hereinafter collectively referred to as the original text) into a vector representation through an encoder, and then use a decoder to decode the vector representation of the original text and translate it into its corresponding translation ( Hereinafter collectively referred to as translations. [0003] In order to obtain better translation results based on machine translation, relevan...

AI Technical Summary

Problems solved by technology

[0003] In order to obtain better translation results based on machine translation, relevant personnel usually post-edit the machine-translated translation, but this method will consume a lot of labor costs

Images