Pushing Paraphrase Away from Original Sentence: A Multi-Round Paraphrase Generation Approach

TL;DR

This paper introduces BTmPG, a multi-round paraphrase generation method that enhances diversity from the original sentence while maintaining semantic integrity, using back-translation for semantic preservation.

Contribution

It proposes a novel multi-round paraphrase generation approach guided by back-translation to improve diversity in neural paraphrase generation.

Findings

BTmPG increases paraphrase diversity significantly.

Semantic preservation is maintained through back-translation.

Both automatic and human evaluations confirm effectiveness.

Abstract

In recent years, neural paraphrase generation based on Seq2Seq has achieved superior performance, however, the generated paraphrase still has the problem of lack of diversity. In this paper, we focus on improving the diversity between the generated paraphrase and the original sentence, i.e., making generated paraphrase different from the original sentence as much as possible. We propose BTmPG (Back-Translation guided multi-round Paraphrase Generation), which leverages multi-round paraphrase generation to improve diversity and employs back-translation to preserve semantic information. We evaluate BTmPG on two benchmark datasets. Both automatic and human evaluation show BTmPG can improve the diversity of paraphrase while preserving the semantics of the original sentence.