SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking

Abstract

In goal-oriented dialog systems, belief trackers estimate the probability distribution of slotvalues at every dialog turn. Previous neural approaches have modeled domain-and slot-dependent belief trackers, and have difficulty in adding new slot-values, resulting in lack of flexibility of domain ontology configurations. In this paper, we propose a new approach to universal and scalable belief tracker, called slot-utterance matching belief tracker (SUMBT). The model learns the relations between domain-slot-types and slotvalues appearing in utterances through attention mechanisms based on contextual semantic vectors. Furthermore, the model predicts slot-value labels in a non-parametric way. From our experiments on two dialog corpora, WOZ 2.0 and MultiWOZ, the proposed model showed performance improvement in comparison with slot-dependent methods and achieved the state-of-the-art joint accuracy.

Recent advances and challenges in task-oriented dialog systems

“…Most recent works adopted belief state for dialog state representation, in which the state is composed of slot-value pairs that represent the user's goal. Therefore, this problem can be formulated as a multi-task classification task [3, [25][26][27]:…”

Conversational Semantic Parsing for Dialog State Tracking

“…Neural classification models (Mrkšić et al, 2017;Mrkšić and Vulić, 2018) alleviate the problem by learning distributed representations of user utterances. However, they still lack scalability to large unbounded output space (Xu and Hu, 2018;Lee et al, 2019) and structured representations. To address the limitations, some recent work treats slot filling as a sequence generation task (Ren et al, 2019;Wu et al, 2019).…”

UniConv: A Unified Conversational Neural Architecture for Multi-domain Task-oriented Dialogues

“…Traditionally, DST models are designed to track states of singledomain dialogues such as WOZ (Wen et al, 2017) and DSTC2 (Henderson et al, 2014a) benchmarks. There have been recent efforts that aim to tackle multi-domain DST such as (Ramadan et al, 2018;Lee et al, 2019;Wu et al, 2019a;. These models can be categorized into two main categories: Fixed vocabulary models (Zhong et al, 2018;Ramadan et al, 2018;Lee et al, 2019), which assume known slot ontology with a fixed candidate set for each slot.…”

“…There have been recent efforts that aim to tackle multi-domain DST such as (Ramadan et al, 2018;Lee et al, 2019;Wu et al, 2019a;. These models can be categorized into two main categories: Fixed vocabulary models (Zhong et al, 2018;Ramadan et al, 2018;Lee et al, 2019), which assume known slot ontology with a fixed candidate set for each slot. On the other hand, open-vocabulary models (Lei et al, 2018;Wu et al, 2019a;Le et al, 2020) derive the candidate set based on the source sequence i.e.…”