MAsked Sequence to Sequence
Year: 2,019
Journal: International Conference on Machine Learning
Languages: Chinese, English, French, German, Romanian
Programming languages: Python
Input data:
words/sentences
Output data:
words, sentences, tokens
Project website: https://github.com/microsoft/MASS
MASS: Masked Sequence to Sequence Pre-training for Language Generation is a novel pre-training method for sequence to sequence based language generation tasks. It randomly masks a sentence fragment in the encoder, and then predicts it in the decoder.