Generative Neural Machine Translation
arXiv:1806.05138
Abstract
We introduce Generative Neural Machine Translation (GNMT), a latent variable architecture which is designed to model the semantics of the source and target sentences. We modify an encoder-decoder translation model by adding a latent variable as a language agnostic representation which is encouraged to learn the meaning of the sentence. GNMT achieves competitive BLEU scores on pure translation tasks, and is superior when there are missing words in the source sentence. We augment the model to facilitate multilingual translation and semi-supervised learning without adding parameters. This framework significantly reduces overfitting when there is limited paired data available, and is effective for translating between pairs of languages not seen during training.
References in corpus (6)
- Auto-Encoding Variational Bayes
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Modeling Coverage for Neural Machine Translation
- TopicRNN: A Recurrent Neural Network with Long-Range Semantic Dependency
- Improved Variational Autoencoders for Text Modeling using Dilated Convolutions
- Bottleneck Conditional Density Estimation
Cited by in corpus (8)
- Transformer-based Conditional Variational Autoencoder for Controllable Story Generation
- Latent-Variable Non-Autoregressive Neural Machine Translation with Deterministic Inference Using a Delta Posterior
- Towards Generating Long and Coherent Text with Multi-Level Latent Variable Models
- Probabilistic Residual Learning for Aleatoric Uncertainty in Image Restoration
- On the Integration of LinguisticFeatures into Statistical and Neural Machine Translation
- Implicit Deep Latent Variable Models for Text Generation
- Neural Machine Translation: A Review and Survey
- Uncertainty-Aware Semantic Augmentation for Neural Machine Translation