Simple and Effective Paraphrastic Similarity from Parallel Translations

09/30/2019

∙

by   John Wieting, et al.

∙

0

∙

share

We present a model and methodology for learning paraphrastic sentence embeddings directly from bitext, removing the time-consuming intermediate step of creating paraphrase corpora. Further, we show that the resulting model can be applied to cross-lingual tasks where it both outperforms and is orders of magnitude faster than more complex state-of-the-art baselines.