Unsupervised Token-wise Alignment to Improve Interpretation of Encoder-Decoder Models

Shun Kiyono, Sho Takase, Jun Suzuki, Naoaki Okazaki, Kentaro Inui, Masaaki Nagata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Developing a method for understanding the inner workings of black-box neural methods is an important research endeavor. Conventionally, many studies have used an attention matrix to interpret how Encoder-Decoder-based models translate a given source sentence to the corresponding target sentence. However, recent studies have empirically revealed that an attention matrix is not optimal for token-wise translation analyses. We propose a method that explicitly models the token-wise alignment between the source and target sequences to provide a better analysis. Experiments show that our method can acquire token-wise alignments that are superior to those of an attention mechanism.

Original languageEnglish
Title of host publicationEMNLP 2018 - 2018 EMNLP Workshop BlackboxNLP
Subtitle of host publicationAnalyzing and Interpreting Neural Networks for NLP, Proceedings of the 1st Workshop
PublisherAssociation for Computational Linguistics (ACL)
Pages74-81
Number of pages8
ISBN (Electronic)9781948087711
Publication statusPublished - 2018
Event1st Workshop on BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, co-located with the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 - Brussels, Belgium
Duration: 2018 Nov 1 → …

Publication series

NameEMNLP 2018 - 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, Proceedings of the 1st Workshop

Conference

Conference1st Workshop on BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, co-located with the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018
Country/TerritoryBelgium
CityBrussels
Period18/11/1 → …

ASJC Scopus subject areas

  • Computer Science Applications
  • Computational Theory and Mathematics
  • Information Systems

Fingerprint

Dive into the research topics of 'Unsupervised Token-wise Alignment to Improve Interpretation of Encoder-Decoder Models'. Together they form a unique fingerprint.

Cite this