SEQ3: Differentiable Sequence-to-Sequence-to-Sequence Autoencoder for Unsupervised Abstractive Sentence Compression

Christos Baziotis, Ion Androutsopoulos, Ioannis Konstas, Alexandros Potamianos

Research output: Contribution to conferencePaper

8 Downloads (Pure)

Abstract

Neural sequence-to-sequence models are currently the dominant approach in
several natural language processing tasks, but require large parallel corpora. We present a sequenceto-sequence-to-sequence autoencoder (SEQ3), consisting of two chained encoder-decoder pairs, with words used as a sequence of discrete latent variables. We apply the proposed model to unsupervised abstractive sentence compression, where the first and last sequences are the input and reconstructed sentences, respectively, while the middle sequence is the compressed sentence. Constraining the length of the latent word sequences forces the model to distill important information from the input. A pretrained language model, acting as a prior over the latent sequences, encourages the compressed sentences to be human-readable. Continuous relaxations enable us to sample from categorical distributions, allowing gradient-based optimization, unlike alternatives that rely on reinforcement learning. The proposed model does not require parallel text-summary pairs, achieving promising results in unsupervised sentence compression on benchmark datasets.
Original languageEnglish
Publication statusPublished - 2019
Event2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics - Minneapolis, United States
Duration: 3 Jun 20195 Jun 2019

Conference

Conference2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics
Abbreviated titleNAACL 2019
CountryUnited States
CityMinneapolis
Period3/06/195/06/19

Fingerprint Dive into the research topics of 'SEQ<sup>3</sup>: Differentiable Sequence-to-Sequence-to-Sequence Autoencoder for Unsupervised Abstractive Sentence Compression'. Together they form a unique fingerprint.

Cite this