D-PAGE: Diverse Paraphrase Generation
This work addresses the need for more diverse paraphrase generation in natural language processing, offering a novel approach that significantly improves diversity metrics, though it is incremental in building upon existing neural machine translation frameworks.
The paper tackles the problem of generating diverse paraphrases by proposing D-PAGE, a method that extends neural machine translation models to produce varied outputs with implicit rewriting patterns, resulting in at least one order of magnitude more diversity than baselines as measured by Jeffrey's Divergence on two benchmark datasets.
In this paper, we investigate the diversity aspect of paraphrase generation. Prior deep learning models employ either decoding methods or add random input noise for varying outputs. We propose a simple method Diverse Paraphrase Generation (D-PAGE), which extends neural machine translation (NMT) models to support the generation of diverse paraphrases with implicit rewriting patterns. Our experimental results on two real-world benchmark datasets demonstrate that our model generates at least one order of magnitude more diverse outputs than the baselines in terms of a new evaluation metric Jeffrey's Divergence. We have also conducted extensive experiments to understand various properties of our model with a focus on diversity.