Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

Abstract : Many modern NLP systems rely on word embeddings, previously trained in an unsupervised manner on large corpora, as base features. Efforts to obtain embeddings for larger chunks of text, such as sentences, have however not been so successful. Several attempts at learning unsupervised representations of sentences have not reached satisfactory enough performance to be widely adopted. In this paper, we show how universal sentence representations trained using the supervised data of the Stanford Natural Language Inference datasets can consistently outperform unsupervised methods like SkipThought vectors on a wide range of transfer tasks. Much like how computer vision uses ImageNet to obtain features, which can then be transferred to other tasks, our work tends to indicate the suitability of natural language inference for transfer learning to other NLP tasks. Our encoder is publicly available.
Type de document :
Pré-publication, Document de travail
EMNLP 2017. 2018
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01897968
Contributeur : Loïc Barrault <>
Soumis le : mercredi 17 octobre 2018 - 18:10:06
Dernière modification le : samedi 20 octobre 2018 - 01:13:40

Lien texte intégral

Identifiants

  • HAL Id : hal-01897968, version 1
  • ARXIV : 1705.02364

Collections

Citation

Alexis Conneau, Douwe Kiela, Holger Schwenk, Loïc Barrault, Antoine Bordes. Supervised Learning of Universal Sentence Representations from Natural Language Inference Data. EMNLP 2017. 2018. 〈hal-01897968〉

Partager

Métriques

Consultations de la notice

21