Schema-Guided Multi-Domain Dialogue State Tracking with Graph Attention Neural Networks

By Lu Chen, Boer Lv, Chi Wang, Su Zhu, Bowen Tan, Kai Yu. 2020

144 citations recorded by the source corpus.

Dialogue state tracking (DST) aims at estimating the current dialogue state given all the preceding conversation. For multi-domain DST, the data sparsity problem is also a major obstacle due to the increased number of state candidates. Existing approaches generally predict the value for each slot independently and do not consider slot relations, which may aggravate the data sparsity problem. In this paper, we propose a Schema-guided multi-domain dialogue State Tracker with graph attention networks (SST) that predicts dialogue states from dialogue utterances and schema graphs which contain slot relations in edges. We also introduce a graph attention matching network to fuse information from utterances and graphs, and a recurrent graph attention network to control state updating. Experiment results show that our approach obtains new state-of-the-art performance on both MultiWOZ 2.0 and MultiWOZ 2.1 benchmarks.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue StateTracking (2021, background): “Current state-of-the-art models (Chen et al., 2020; Lin et al., 2020; Heck et al., 2020; Hosseini-Asl et al., 2020; Ye et al., 2021; Li et al., 2020) trained with extensive annotated data have been shown promising performance in complex multi-domain conversations (Budzianowski et al.”
A General Survey on Attention Mechanisms in Deep Learning (2022, methodology): “In [69], multi-dimensional attention is used to extend graph attention networks for dialogue state tracking.”
Dialogue State Tracking with a Language Model using Schema-Driven Prompting (2021, methodology): “, 2020a), SST (Chen et al., 2020), TripPy (Heck et al.”
MinTL: Minimalist Transfer Learning for Task-Oriented Dialogue Systems (2020, methodology): “It is important to decouple the two de-coders because the distributions between the Lev decoder and response decoder are different. model SST (Chen et al., 2020), our model obtains a 1.62% lower joint goal accuracy on MultiWOZ 2.1.”
Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialogue State Tracking (2021, methodology): “Some previous work utilized the structure via an extra schema graph in a regular training process (Chen et al., 2020; Zhu et al., 2020; Wu et al., 2020).”
GODEL: Large-Scale Pre-Training for Goal-Directed Dialog (2022, background): “• The Schema-Guided Dialog (Chen et al., 2020) is a multi-domain, task-oriented conversation dataset that contains approximately 20k annotated examples.”
Find or Classify? Dual Strategy for Slot-Value Predictions on Multi-Domain Dialog State Tracking (2019, background): “into two classes (Xu and Hu,2018;Gao et al.,2019b;Ramadan et al., 2018;Zhong et al.,2018): i.e., ontology-based and ontology-free. The ontology-based approaches (Ramadan et al.,2018;Zhong et al.,2018;Chen et al., 2020) require full access to the pre-deﬁned ontology to perform classiﬁcation over the ”