Pervasive Attention: 2D Convolutional Neural Networks for Sequence-to-Sequence Prediction

Abstract : Current state-of-the-art machine translation systems are based on encoder-decoder architectures, that first encode the input sequence, and then generate an output sequence based on the input encoding. Both are interfaced with an attention mechanism that recombines a fixed encoding of the source tokens based on the decoder state. We propose an alternative approach which instead relies on a single 2D convolutional neural network across both sequences. Each layer of our network re-codes source tokens on the basis of the output sequence produced so far. Attention-like properties are therefore pervasive throughout the network. Our model yields excellent results, outperforming state-of-the-art encoder-decoder systems, while being conceptually simpler and having fewer parameters.
Type de document :
Communication dans un congrès
Conference on Computational Natural Language Learning, 2018, Brussels, Belgium
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01851612
Contributeur : Thoth Team <>
Soumis le : vendredi 10 août 2018 - 14:05:45
Dernière modification le : vendredi 10 août 2018 - 16:37:05

Identifiants

  • HAL Id : hal-01851612, version 1

Collections

Citation

Maha Elbayad, Laurent Besacier, Jakob Verbeek. Pervasive Attention: 2D Convolutional Neural Networks for Sequence-to-Sequence Prediction. Conference on Computational Natural Language Learning, 2018, Brussels, Belgium. 〈hal-01851612〉

Partager

Métriques

Consultations de la notice

217

Téléchargements de fichiers

39