arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2002.08898cs.CLcs.AIcs.LGstat.ML
收藏

MA-DST:基于多注意力的可扩展对话状态跟踪

MA-DST: Multi-Attention Based Scalable Dialog State Tracking

University of Wisconsin-Madison(威斯康星大学麦迪逊分校);Amazon Alexa AI(亚马逊 Alexa 人工智能)

Adarsh Kumar, Peter Ku, Anuj Kumar Goyal, Angeliki Metallinou, Dilek Hakkani-Tur

摘要:

面向任务的对话代理为用户提供了一个自然语言接口来完成他们的任务。对话状态跟踪(DST)通常是这些系统的核心组件,它跟踪系统在整个对话中对用户目标的理解。为了实现准确的多域DST,该模型需要对过去的话语和槽语义之间的依赖关系进行编码,并理解对话上下文,包括远程跨域引用。我们引入了一种新的架构,这个任务的会话历史和插槽语义编码更强大的使用注意力机制在多个粒度。特别是,我们使用交叉注意力模型的上下文和插槽之间的关系在不同的语义水平和自我注意力来解决跨域共指。此外,我们提出的架构不依赖于事先知道的域本体,也可以用于新的域或看不见的槽值的zero-shot设置。与MultiWoZ 2.1数据集上目前最先进的技术相比,我们的模型在全数据设置中将联合目标准确度提高了5%(绝对),在zero-shot设置中提高了2%(绝对)。

英文摘要:

Task oriented dialog agents provide a natural language interface for users to complete their goal. Dialog State Tracking (DST), which is often a core component of these systems, tracks the system's understanding of the user's goal throughout the conversation. To enable accurate multi-domain DST, the model needs to encode dependencies between past utterances and slot semantics and understand the dialog context, including long-range cross-domain references. We introduce a novel architecture for this task to encode the conversation history and slot semantics more robustly by using attention mechanisms at multiple granularities. In particular, we use cross-attention to model relationships between the context and slots at different semantic levels and self-attention to resolve cross-domain coreferences. In addition, our proposed architecture does not rely on knowing the domain ontologies beforehand and can also be used in a zero-shot setting for new domains or unseen slot values. Our model improves the joint goal accuracy by 5% (absolute) in the full-data setting and by up to 2% (absolute) in the zero-shot setting over the present state-of-the-art on the MultiWoZ 2.1 dataset.

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。

↑