Abstract
Data-to-text generation can be conceptually divided into two parts: ordering and structuring the information (planning), and generating fluent language describing the information (realization). Modern neural generation systems conflate these two steps into a single end-to-end differentiable system. We propose to split the generation process into a symbolic text-planning stage that is faithful to the input, followed by a neural generation stage that focuses only on realization. For training a plan-to-text generator, we present a method for matching reference texts to their corresponding text plans. For inference time, we describe a method for selecting high-quality text plans for new inputs. We implement and evaluate our approach on the WebNLG benchmark. Our results demonstrate that decoupling text planning from neural realization indeed improves the system's reliability and adequacy while maintaining fluent output. We observe improvements both in BLEU scores and in manual evaluations. Another benefit of our approach is the ability to output diverse realizations of the same input, paving the way to explicit control over the generated text structure.
Abstract (translated)
从概念上讲,数据到文本的生成可以分为两部分:排序和构造信息(规划),生成描述信息的流畅语言(实现)。现代神经生成系统将这两个步骤合并为一个端到端的可微分系统。我们建议将生成过程分成一个忠实于输入的符号文本规划阶段,然后是一个只关注实现的神经生成阶段。为了训练计划到文本生成器,我们提出了一种将参考文本与其对应的文本计划匹配的方法。对于推理时间,我们描述了一种为新输入选择高质量文本计划的方法。我们在WebNLG基准上实施和评估我们的方法。结果表明,将文本规划与神经网络实现相分离,在保证输出流畅的同时,确实提高了系统的可靠性和充分性。我们观察到布鲁评分和手工评估的改善。我们的方法的另一个好处是能够输出相同输入的不同实现,为显式控制生成的文本结构铺平了道路。
URL
https://arxiv.org/abs/1904.03396