Discourse-Based Objectives for Fast Unsupervised Sentence Representation Learning

Abstract

This work presents a novel objective function for the unsupervised training of neural network sentence encoders. It exploits signals from paragraph-level discourse coherence to train these models to understand text. Our objective is purely discriminative, allowing us to train models many times faster than was possible under prior methods, and it yields models which perform well in extrinsic evaluations.

Publication
arXiv preprint arXiv:1705.00557
Yacine Jernite
Yacine Jernite
PhD student

Research Scientist, Hugging Face

David Sontag
David Sontag
Associate Professor of EECS

My research focuses on advancing machine learning and artificial intelligence, and using these to transform health care.

Related