Find or Classify? Dual Strategy for Slot-Value Predictions on Multi-Domain Dialog State Tracking

Zhang, Jianguo; Hashimoto, Kazuma; Wu, Chien-Sheng; Wan, Yao; Yu, Philip S.; Socher, Richard; Xiong, Caiming

doi:10.48550/arxiv.1910.03544

Cited by 31 publications

(62 citation statements)

References 23 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Intent Classification Intent classification aims to identify the intents of user utterances, which is a critical component of goal-oriented dialog systems. Attaining high intent classification accuracy is an important step towards solving many downstream tasks such as dialogue state tracking Zhang et al, 2019) and dialogue management (Gao et al, 2018;Ham et al, 2020). A practical challenge is data scarcity because different systems define different sets of intents, and it is costly to obtain enough utterance samples for each intent.…”

Section: Semantic Textual Similaritymentioning

confidence: 99%

Virtual Augmentation Supported Contrastive Learning of Sentence Representations

Zhang¹,

Xiao²,

Zhu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Despite profound successes, contrastive representation learning relies on carefully designed data augmentations using domain specific knowledge. This challenge is magnified in natural language processing where no general rules exist for data augmentation due to the discrete nature of natural language. We tackle this challenge by presenting a Virtual augmentation Supported Contrastive Learning of sentence representations (VaSCL). Originating from the interpretation that data augmentation essentially constructs the neighborhoods of each training instance, we in turn utilize the neighborhood to generate effective data augmentations. Leveraging the large training batch size of contrastive learning, we approximate the neighborhood of an instance via it's K-nearest in-batch neighbors in the representation space. We then define an instance discrimination task within this neighborhood, and generate the virtual augmentation in an adversarial training manner. We access the performance of VaSCL 1 on a wide range of downstream tasks, and set a new state-of-the-art for unsupervised sentence representation learning.

show abstract

Section: Semantic Textual Similaritymentioning

confidence: 99%

Virtual Augmentation Supported Contrastive Learning of Sentence Representations

Zhang¹,

Xiao²,

Zhu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…, 2007;Williams et al, 2014). Recent state-of-the-art models (Lei et al, 2018;Wu et al, 2020;Peng et al, 2020;Zhang et al, 2019;Heck et al, 2020;Mehri et al, 2020;Hosseini-Asl et al, 2020; trained with extensive annotated dialogue data have shown promising performance in complex multi-domain conversations (Budzianowski et al, 2018;. However, collecting large amounts of data for every dialogue domain is often costly and inefficient.…”

Section: Related Workmentioning

confidence: 99%

Zero-Shot Dialogue State Tracking via Cross-Task Transfer

Lin¹,

Liu²,

Madotto³

et al. 2021

Preprint

View full text Add to dashboard Cite

Zero-shot transfer learning for dialogue state tracking (DST) enables us to handle a variety of task-oriented dialogue domains without the expense of collecting in-domain data. In this work, we propose to transfer the crosstask knowledge from general question answering (QA) corpora for the zero-shot DST task. Specifically, we propose TransferQA, a transferable generative QA model that seamlessly combines extractive QA and multichoice QA via a text-to-text transformer framework, and tracks both categorical slots and non-categorical slots in DST. In addition, we introduce two effective ways to construct unanswerable questions, namely, negative question sampling and context truncation, which enable our model to handle "none" value slots in the zero-shot DST setting. The extensive experiments show that our approaches substantially improve the existing zero-shot and few-shot results on MultiWoz. Moreover, compared to the fully trained baseline on the Schema-Guided Dialogue dataset, our approach shows better generalization ability in unseen domains.

show abstract

“…Slot Value. Following recent works (Zhang et al, 2019;, slots are divided into categorical and non-categorical slots. For categorical slots, we incorporate the candidate values into the slot description, i.e., "[slot] of the [domain] is [value-1] or [value-2]?".…”

Section: Slot Description Variantsmentioning

confidence: 99%

Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue State Tracking

Lin¹,

Liu²,

Moon³

et al. 2021

Preprint

View full text Add to dashboard Cite

Zero-shot cross-domain dialogue state tracking (DST) enables us to handle task-oriented dialogue in unseen domains without the expense of collecting in-domain data.In this paper, we propose a slot description enhanced generative approach for zero-shot cross-domain DST. Specifically, our model first encodes dialogue context and slots with a pre-trained self-attentive encoder, and generates slot values in an auto-regressive manner. In addition, we incorporate Slot Type Informed Descriptions that capture the shared information across slots to facilitate cross-domain knowledge transfer. Experimental results on the MultiWOZ dataset show that our proposed method significantly improves existing stateof-the-art results in the zero-shot cross-domain setting.

show abstract

Find or Classify? Dual Strategy for Slot-Value Predictions on Multi-Domain Dialog State Tracking

Cited by 31 publications

References 23 publications

Virtual Augmentation Supported Contrastive Learning of Sentence Representations

Virtual Augmentation Supported Contrastive Learning of Sentence Representations

Zero-Shot Dialogue State Tracking via Cross-Task Transfer

Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue State Tracking

Contact Info

Product

Resources

About