%0 Journal Article
%T Skip-Thought Vectors
%A Ryan Kiros
%A Yukun Zhu
%A Ruslan Salakhutdinov
%A Richard S. Zemel
%A Antonio Torralba
%A Raquel Urtasun
%A Sanja Fidler
%J Computer Science
%D 2015
%I arXiv
%X 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.
%U http://arxiv.org/abs/1506.06726v1