Skip-Thought Vectors

TL;DR

This paper introduces an unsupervised method for learning universal sentence embeddings using a context-based training approach, resulting in robust representations that perform well across various NLP tasks.

Contribution

The authors propose a novel unsupervised training method for sentence encoding based on text continuity, and introduce a vocabulary expansion technique for broader word coverage.

Findings

High performance on semantic relatedness and paraphrase detection tasks

Effective in image-sentence ranking and question classification

Robust sentence representations suitable for multiple NLP applications

Abstract

We describe an approach for unsupervised learning of a generic, distributed sentence encoder. Using the continuity of text from books, we train an encoder-decoder model that tries to reconstruct the surrounding sentences of an encoded passage. Sentences that share semantic and syntactic properties are thus mapped to similar vector representations. We next introduce a simple vocabulary expansion method to encode words that were not seen as part of training, allowing us to expand our vocabulary to a million words. After training our model, we extract and evaluate our vectors with linear models on 8 tasks: semantic relatedness, paraphrase detection, image-sentence ranking, question-type classification and 4 benchmark sentiment and subjectivity datasets. The end result is an off-the-shelf encoder that can produce highly generic sentence representations that are robust and perform well in practice. We will make our encoder publicly available.