Toward Scalable Neural Dialogue State Tracking Model

By Elnaz Nouri, Ehsan Hosseini-Asl. 2018

84 citations recorded by the source corpus.

The latency in the current neural based dialogue state tracking models prohibits them from being used efficiently for deployment in production systems, albeit their highly accurate performance. This paper proposes a new scalable and accurate neural dialogue state tracking model, based on the recently proposed Global-Local Self-Attention encoder (GLAD) model by Zhong et al. which uses global modules to share parameters between estimators for different types (called slots) of dialogue states, and uses local modules to learn slot-specific features. By using only one recurrent networks with global conditioning, compared to (1 + \# slots) recurrent networks with global and local conditioning used in the GLAD model, our proposed model reduces the latency in training and inference times by $35\%$ on average, while preserving performance of belief state tracking, by $97.38\%$ on turn request and $88.51\%$ on joint goal and accuracy. Evaluation on Multi-domain dataset (Multi-WoZ) also demonstrates that our model outperforms GLAD on turn inform and joint goal accuracy.

Subfields: Artificial Intelligence, Natural Language Processing

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

In Natural Language Processing

Citations in the corpus

A Simple Language Model for Task-Oriented Dialogue (2020, background): “These components include understanding user intent via intent detection [26], tracking the constraints imposed by the user via dialogue state tracking [20, 32, 41, 34, 53, 59, 64, 8, 19], determining system actions via dialogue policy [51], and using dedicated response generation components [49].”
Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems (2019, background): “Note that the MDBT, GLAD, and GCE models each need a pre-deﬁned domain ontology to perform binary clas-siﬁcation for each ontology term, which hinders their DST tracking performance, as mentioned in Section 1.”
MultiWOZ 2. (2019, background): “…a unique resource, in terms of its multi-domain interactions as well as slot value transfers be-tween these domains, and has quickly attracted researchers for dialogue state tracking (Nouri and Hosseini-Asl, 2018; Goel et al., 2019; Wu et al., 2019) and dialogue policy learning (Zhao et al., 2019).”
Multi-domain Dialogue State Tracking as Dynamic Knowledge Graph Enhanced Question Answering (2019, methodology): “For example, Nouri & Hosseini-Asl (2018) improves Zhong et al. (2018)’s work by removing local models and building a slot-conditioned global model to share parameters between domains and slots, thus computing a score for every (domain, slot, value) tuple.”
SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking (2019, methodology): “, 2018) or have difficulty in adding new slot-values that are not defined in the ontology (Ramadan et al., 2018; Nouri and Hosseini-Asl, 2018).”
MultiWOZ 2.1: Multi-Domain Dialogue State Corrections and State Tracking Baselines (2019, background): “This dataset has been a unique resource, in terms of the multi-domain interactions as well as slot value transfers between these domains, and has quickly attracted researchers for dialogue state tracking (Nouri and Hosseini-Asl, 2018; Goel et al., 2019; Wu et al., 2019) and dialogue policy learning ”
Dialog State Tracking: A Neural Reading Comprehension Approach (2019, methodology): “First, there is Method Accuracy MultiWOZ Benchmark 25.83% GLAD (Zhong et al., 2018) 35.57% GCE (Nouri and Hosseini-Asl, 2018) 35.58% Our approach (single) 39.41% Our approach (ensemble) 42.12% HyST (ensemble) (Goel et al., 2019) 44.22% Our approach + JST (ensemble) 47.33 % lack of consistency betwee”
Schema-Guided Multi-Domain Dialogue State Tracking with Graph Attention Neural Networks (2020, methodology): “The classification-based ones include: MDBT (Ramadan, Budzianowski, and Gasic 2018), which uses multiple bi-LSTMs to encode utterances of system and user; GLAD (Zhong, Xiong, and Socher 2018), a global-local self-attention model; GCE (Nouri and Hosseini-Asl 2018), a global-conditioned encoder; HJST ”
Dialog State Tracking with Reinforced Data Augmentation (2019, methodology): “Following the previous work (Ren et al., 2018; Zhong et al., 2018; Nouri and Hosseini-Asl, 2018), we concatenate the pre-trained GloVe embed-dings (Pennington et al., 2014) and the character embeddings (Hashimoto et al., 2017) as the ﬁnal word embeddings and keep them ﬁxed when training.”
MinTL: Minimalist Transfer Learning for Task-Oriented Dialogue Systems (2020, methodology): “The former includes MDBT (Ra-madan et al., 2018), GLAD (Zhong et al., 2018), GCE (Nouri and Hosseini, 2018), FJST (Eric et al., 2019), HyST (Goel et al., 2019), SUMBT (Lee et al., 2019), SST (Chen et al., 2020), TOD-BERT (Wu et al., 2020), and DST-Picklist (Zhang et al., 2019a); the latter includes ”