arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.07499cs.CLcs.AI
收藏

一种联合自然语言理解和生成的生成模型

A Generative Model for Joint Natural Language Understanding and Generation

University of Cambridge(剑桥大学);Apple(苹果公司)

Bo-Hsiang Tseng, Jianpeng Cheng, Yimai Fang, David Vandyke

摘要:

自然语言理解(NLU)和自然语言生成(NLG)是构建面向任务的对话系统的两个基本且相关的任务,它们具有相反的目标:NLU处理从自然语言到形式表示的转换,而NLG则相反。在任一任务中成功的关键是并行训练数据,其大规模获得是昂贵的。在这项工作中,我们提出了一个生成模型,该模型通过共享潜变量将NLU和NLG耦合起来。这种方法允许我们探索自然语言和形式表示的空间,并通过潜在空间促进信息共享,最终使NLU和NLG受益。我们的模型在两个对话数据集上实现了最先进的性能,具有平面和树结构的正式表示。我们还表明,该模型可以通过利用未标记的数据以半监督的方式进行训练,以提高其性能。

英文摘要:

Natural language understanding (NLU) and natural language generation (NLG) are two fundamental and related tasks in building task-oriented dialogue systems with opposite objectives: NLU tackles the transformation from natural language to formal representations, whereas NLG does the reverse. A key to success in either task is parallel training data which is expensive to obtain at a large scale. In this work, we propose a generative model which couples NLU and NLG through a shared latent variable. This approach allows us to explore both spaces of natural language and formal representations, and facilitates information sharing through the latent space to eventually benefit NLU and NLG. Our model achieves state-of-the-art performance on two dialogue datasets with both flat and tree-structured formal representations. We also show that the model can be trained in a semi-supervised fashion by utilising unlabelled data to boost its performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑