Advancing Seq2seq with Joint Paraphrase Learning

Abstract

We address the problem of model generalization for sequence to sequence (seq2seq) architectures. We propose going beyond data augmentation via paraphrase-optimized multi-task learning and observe that it is useful in correctly handling unseen sentential paraphrases as inputs. Our models greatly outperform SOTA seq2seq models for semantic parsing on diverse domains (Overnight -up to 3.2% and emrQA -7%) and Nematus (Sennrich et al., 2017), the winning solution for WMT 2017, for Czech to English translation (CzENG 1.6 -1.5 BLEU). * Work done while So Yeon Min was a student at MIT. 1 AllenNLP's competitive BiDAF (Seo et al.