Recent developments in neural networks have led to the advance in data-to-text generation. However, the lack of ability of neural models to control the structure of generated output can be limiting in certain real-world applications. In this study, we propose a novel Plan-then-Generate (PlanGen) framework to improve the controllability of neural data-to-text models. Extensive experiments and analyses are conducted on two benchmark datasets, ToTTo and WebNLG. The results show that our model is able to control both the intra-sentence and inter-sentence structure of the generated output. Furthermore, empirical comparisons against previous state-of-the-art methods show that our model improves the generation quality as well as the output diversity as judged by human and automatic evaluations.
@article{arxiv.2108.13740,
title = {Plan-then-Generate: Controlled Data-to-Text Generation via Planning},
author = {Yixuan Su and David Vandyke and Sihui Wang and Yimai Fang and Nigel Collier},
journal= {arXiv preprint arXiv:2108.13740},
year = {2021}
}