Computer Science
4 years ago

New language model boosts machine translation accuracy for low-resource languages!

International Journal of Artificial Intelligence Tools
Language Model Pre-training Method in Machine Translation Based on Named Entity Recognition
Zhen Li, Dan Qu, Chaojie Xie, Wenlin Zhang, Yanxia Li

Paper Summary

Related papers

Theme background image
Computer Science
4 years ago

New language model boosts machine translation accuracy for low-resource languages!

Machine Translation
Natural Language Processing
1 view
Theme background image
Computer Science
8 years ago

Automated post-editor boosts statistical machine translation accuracy without linguistic tools.

Machine Translation
Natural Language Processing
Text Compression
Theme background image
Computer Science
8 years ago

New measure improves machine translation accuracy using sentence structure information.

Machine Translation
Rough Sets

The article discusses a new method for improving machine translation for languages with limited resources. By using a technique called named entity recognition masked language model, the researchers were able to train a language model that better understands the meaning of words in a sentence. This method resulted in the highest translation accuracy reported so far for unsupervised machine translation tasks.