Curate and Generate: A Corpus and Method for Joint Control of Semantics and Style in Neural NLG
Curate and Generate: A Corpus and Method for Joint Control of Semantics and Style in Neural NLG
复制标题
DOI：
10.18653/v1/p19-1596
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Shereen Oraby;Vrindavan Harrison;Abteen Ebrahimi;M. Walker
中科院分区：
文献类型：
--
作者：
Shereen Oraby;Vrindavan Harrison;Abteen Ebrahimi;M. Walker
基于结构化语义表示的神经自然语言生成（NNLG）近年来越来越受欢迎。虽然我们已经看到在生成保留语义的语法正确的话语方面取得了进展，但NNLG系统的各种缺点也很明显：新任务需要新的训练数据，这些数据不可用或无法直接获取，模型输出简单，可能是枯燥和重复的。本文通过以下方式解决了NNLG中的这两个关键挑战：（1）通过使用来自免费提供的自然描述性用户评论的数据，可扩展地（且无成本地）创建具有丰富样式标记的并行含义表示和参考文本的训练数据集，以及（2）系统地探索样式标记如何实现对神经模型输出的语义和风格方面的联合控制。我们提出了YelpNLG，一个包含30万个丰富的平行意义表示和跨越不同餐厅属性的高度风格多样的参考文本的语料库，并描述了一种新的方法，该方法可以可扩展地重用于为其他领域生成NLG数据集。实验表明，模型控制了重要的方面，包括形容词的词汇选择，输出长度和情感，使模型能够成功地击中多个风格目标，而不牺牲语义。
Neural natural language generation (NNLG) from structured meaning representations has become increasingly popular in recent years. While we have seen progress with generating syntactically correct utterances that preserve semantics, various shortcomings of NNLG systems are clear: new tasks require new training data which is not available or straightforward to acquire, and model outputs are simple and may be dull and repetitive. This paper addresses these two critical challenges in NNLG by: (1) scalably (and at no cost) creating training datasets of parallel meaning representations and reference texts with rich style markup by using data from freely available and naturally descriptive user reviews, and (2) systematically exploring how the style markup enables joint control of semantic and stylistic aspects of neural model output. We present YelpNLG, a corpus of 300,000 rich, parallel meaning representations and highly stylistically varied reference texts spanning different restaurant attributes, and describe a novel methodology that can be scalably reused to generate NLG datasets for other domains. The experiments show that the models control important aspects, including lexical choice of adjectives, output length, and sentiment, allowing the models to successfully hit multiple style targets without sacrificing semantics.