Text correction model training method and device and text correction method and device

Abstract

The invention discloses a text correction model training method and device and a text correction method and device.The text correction method and device are used for improving text correction accuracy, and the training method comprises the steps of inputting wrong texts into a text correction model to acquire a correction word vector sequence; based on the correction word vector sequence and the target word vector sequence, obtaining context semantic information representing text fluency corresponding to each correction word vector; based on the context semantic information corresponding to each correction word vector, respectively obtaining a generation probability corresponding to each correction word vector, the generation probability being a probability of generating the correction word vector with the context semantic information; obtaining a first loss value based on the generation probability of each correction word vector and the difference degree of each target word vector; obtaining a second loss value based on the difference degree between the target word vector sequence and the correction word vector sequence; and updating parameters of the text correction model based on a weighted summation result of the first loss value and the second loss value.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a text correction model training method and device, and a text correction method and device. Background technique [0002] This section is intended to provide a background or context to the implementations of the application that are recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] At present, common text correction models (such as English composition grammar correction models) all use an end-to-end model framework based on Transformer (machine translation), which includes two parts: Encoder (encoder) and Decoder (decoder). . In the process of model training, the target distribution is often used as 0-1 distribution, that is, only the probability corresponding to the only correct target is used when calculating the loss, which will cause the model to fail to learn the diversity of tr...

AI Technical Summary

Problems solved by technology

In the process of model training, the target distribution is often used as 0-1 distribution, that is, only the probability corresponding to the only correct target is used when calculating the loss, which will cause the model to fail to learn the diversity of training data well and reduce the accuracy of the model. output accuracy

Images