Learning Bilingual Sentence Embeddings via Autoencoding and Computing Similarities with a Multilayer Perceptron
arXiv:1906.01942
Abstract
We propose a novel model architecture and training algorithm to learn bilingual sentence embeddings from a combination of parallel and monolingual data. Our method connects autoencoding and neural machine translation to force the source and target sentence embeddings to share the same space without the help of a pivot language or an additional transformation. We train a multilayer perceptron on top of the sentence embeddings to extract good bilingual sentence pairs from nonparallel or noisy parallel data. Our approach shows promising performance on sentence alignment recovery and the WMT 2018 parallel corpus filtering tasks with only a single model.
ACL 2019 Repl4NLP camera-ready
References in corpus (7)
- Scikit-learn: Machine Learning in Python
- Achieving Human Parity on Automatic Chinese to English News Translation
- A Hierarchical Neural Autoencoder for Paragraphs and Documents
- Sockeye: A Toolkit for Neural Machine Translation
- Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond
- The Missing Ingredient in Zero-Shot Neural Machine Translation
- An Empirical Analysis of NMT-Derived Interlingual Embeddings and their Use in Parallel Sentence Identification