Syntax-guided Controlled Generation of Paraphrases
Preprint
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 availableKeywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Latent Part-of-Speech Sequences for Neural Machine Translation2019
- A Structured Review of the Validity of BLEUComputational Linguistics, 2018
- Paraphrase Generation with Deep Reinforcement Learning2018
- Adversarial Example Generation with Syntactically Controlled Paraphrase Networks2018
- Does String-Based Neural MT Learn Source Syntax?2016
- Generating Sentences from a Continuous Space2016
- A survey of research on text simplificationITL - International Journal of Applied Linguistics, 2014
- Generating Phrasal and Sentential Paraphrases: A Survey of Data-Driven MethodsComputational Linguistics, 2010
- The Roots of Coherence in DiscourseHuman Development, 2003
- Simple Fast Algorithms for the Editing Distance between Trees and Related ProblemsSIAM Journal on Computing, 1989