2018/05/01 by Zheng Zhang, Minlie Huang, Zhang, Zheng +9 · 5 citations
Computer Science · #68T50 #Artificial Intelligence (cs.AI) #Artificial intelligence #Cognitive psychology #Computation and Language (cs.CL) #Computer science #Context (archaeology) #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Intelligent Tutoring Systems and Adaptive Learning #Memorization #Memory management #Natural language processing #Programming language #Range (aeronautics) #Representation (politics) #Semantics (computer science) #Session (web analytics) #Speech and dialogue systems #State (computer science) #Task (project management) #Topic Modeling #World Wide Web #cs.AI #cs.CL #cs.IR #msc:68T50
paper · pdf · doi:10.48550/arxiv.1805.00150
published in arXiv (Cornell University) (Cornell University) · 25 pages, 9 figures, Under review of ACM Transactions on Information Systems (TOIS)
arxiv created 2018/05/01 · openalex publication_date 2018/05/01 · arxiv updated 2018/05/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
Dialogue management (DM) decides the next action of a dialogue system according to the current dialogue state, and thus plays a central role in task-oriented dialogue systems. Since dialogue management requires to have access to not only local utterances, but also the global semantics of the entire dialogue session, modeling the long-range history information is a critical issue. To this end, we propose a novel Memory-Augmented Dialogue management model (MAD) which employs a memory controller and two additional memory structures, i.e., a slot-value memory and an external memory. The slot-value memory tracks the dialogue state by memorizing and updating the values of semantic slots (for instance, cuisine, price, and location), and the external memory augments the representation of hidden states of traditional recurrent neural networks through storing more context information. To update the dialogue state efficiently, we also propose slot-level attention on user utterances to extract specific semantic information for each slot. Experiments show that our model can obtain state-of-the-art performance and outperforms existing baselines.