Dialogue systems have made massive promising progress contributed by deep learning techniques and have been widely applied in our life. However, existing end-to-end neural models suffer from the problem of tending to generate uninformative and generic responses because they cannot ground dialogue context with background knowledge. In order to solve this problem, many researchers begin to consider combining external knowledge in dialogue systems, namely knowledge-enhanced dialogue systems. The challenges of knowledge-enhanced dialogue systems include how to select the appropriate knowledge from large-scale knowledge bases, how to read and understand extracted knowledge, and how to integrate knowledge into responses generation process. Combined with external knowledge, dialogue systems can deeply understand the dialogue context, and generate more informative and logical responses. This survey gives a comprehensive review of knowledge-enhanced dialogue systems, summarizes research progress to solve these challenges and proposes some open issues and research directions.

对话系统在深度学习技术的推动下取得了巨大的进步，并已广泛应用于我们的生活中。然而，现有的端到端神经模型存在倾向于生成无信息和通用响应的问题，因为它们无法将对话上下文与背景知识结合起来。为了解决这个问题，很多研究者开始考虑在对话系统中结合外部知识，即知识增强型对话系统。知识增强对话系统的挑战包括如何从大规模知识库中选择合适的知识，如何阅读和理解提取的知识，以及如何将知识整合到响应生成过程中。结合外部知识，对话系统可以深入理解对话上下文，并生成更多信息和逻辑响应。本次调查全面回顾了知识增强对话系统，总结了解决这些挑战的研究进展，并提出了一些开放性问题和研究方向。