Dynamic Schema Graph Fusion Network for Multi-Domain Dialogue State Tracking
This addresses a domain-specific problem in DST for improving generalization to unseen domains, with incremental contributions.
The paper tackles the problem of modeling relations among domains and slots in Dialogue State Tracking (DST) by proposing DSGFNet, which fuses prior slot-domain membership relations and dialogue-aware dynamic slot relations to improve generalization to unseen domains. Empirical results show DSGFNet outperforms existing methods on benchmark datasets SGD, MultiWOZ2.1, and MultiWOZ2.2.
Dialogue State Tracking (DST) aims to keep track of users' intentions during the course of a conversation. In DST, modelling the relations among domains and slots is still an under-studied problem. Existing approaches that have considered such relations generally fall short in: (1) fusing prior slot-domain membership relations and dialogue-aware dynamic slot relations explicitly, and (2) generalizing to unseen domains. To address these issues, we propose a novel \textbf{D}ynamic \textbf{S}chema \textbf{G}raph \textbf{F}usion \textbf{Net}work (\textbf{DSGFNet}), which generates a dynamic schema graph to explicitly fuse the prior slot-domain membership relations and dialogue-aware dynamic slot relations. It also uses the schemata to facilitate knowledge transfer to new domains. DSGFNet consists of a dialogue utterance encoder, a schema graph encoder, a dialogue-aware schema graph evolving network, and a schema graph enhanced dialogue state decoder. Empirical results on benchmark datasets (i.e., SGD, MultiWOZ2.1, and MultiWOZ2.2), show that DSGFNet outperforms existing methods.