Syntax-guided Controlled Generation of Paraphrases

TL;DR

This paper introduces SGCP, a novel end-to-end framework for controlled paraphrase generation guided by complex syntactic constraints, outperforming existing methods in relevance and syntactic conformity.

Contribution

The paper proposes SGCP, a new method that incorporates detailed syntactic guidance from exemplars for improved controlled paraphrase generation.

Findings

SGCP generates syntactically conforming paraphrases effectively.

SGCP outperforms state-of-the-art baselines in automated and human evaluations.

The approach maintains relevance while adhering to syntactic constraints.

Abstract

Given a sentence (e.g., "I like mangoes") and a constraint (e.g., sentiment flip), the goal of controlled text generation is to produce a sentence that adapts the input sentence to meet the requirements of the constraint (e.g., "I hate mangoes"). Going beyond such simple constraints, recent works have started exploring the incorporation of complex syntactic-guidance as constraints in the task of controlled paraphrase generation. In these methods, syntactic-guidance is sourced from a separate exemplar sentence. However, these prior works have only utilized limited syntactic information available in the parse tree of the exemplar sentence. We address this limitation in the paper and propose Syntax Guided Controlled Paraphraser (SGCP), an end-to-end framework for syntactic paraphrase generation. We find that SGCP can generate syntax conforming sentences while not compromising on relevance. We perform extensive automated and human evaluations over multiple real-world English language datasets to demonstrate the efficacy of SGCP over state-of-the-art baselines. To drive future research, we have made SGCP's source code available