Factorising Meaning and Form for Intent-Preserving Paraphrasing

ACL 2021  ·  Tom Hosking, Mirella Lapata ·

We propose a method for generating paraphrases of English questions that retain the original intent but use a different surface form. Our model combines a careful choice of training objective with a principled information bottleneck, to induce a latent encoding space that disentangles meaning and form... We train an encoder-decoder model to reconstruct a question from a paraphrase with the same meaning and an exemplar with the same surface form, leading to separated encoding spaces. We use a Vector-Quantized Variational Autoencoder to represent the surface form as a set of discrete latent variables, allowing us to use a classifier to select a different surface form at test time. Crucially, our method does not require access to an external source of target exemplars. Extensive experiments and a human evaluation show that we are able to generate paraphrases with a better tradeoff between semantic preservation and syntactic novelty compared to previous methods. read more

PDF Abstract ACL 2021 PDF ACL 2021 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Paraphrase Generation Paralex Separator iBLEU 14.84 # 1
Paraphrase Generation Quora Question Pairs Separator iBLEU 5.84 # 1