paper-with-me

Papers

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

2021-05-10 · Zhaojiang Lin, Bing Liu, Seungwhan Moon, Paul Crook, Zhenpeng Zhou, Zhiguang Wang, Zhou Yu, Andrea Madotto, Eunjoon Cho, Rajen Subba

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 state-of-the-art results in the zero-shot cross-domain setting.

📄 PDF Abstract BibTeX arXiv:2105.04222

Code (2)

facebookresearch/Zero-Shot-DST 공식 구현 pytorch
devjwsong/t5-dst-modified-pytorch pytorch

Tasks

Dialogue State TrackingTransfer Learning

Methods 이 논문이 사용한 방법론

DST Dynamic sparse training methods train neural networks in a sparse manner, starting with an initial sparse mask, and periodically updating the mask based on some criteria.

Similar Papers 제목 키워드 기반

Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue StateTracking

2021-06-01 · NAACL 2021 4 · Zhaojiang Lin, Bing Liu, Seungwhan Moon, Paul Crook 외

Zero-shot cross-domain dialogue state tracking (DST) enables us to handle unseen domains without the expense of collecting in-domain data. In this paper, we propose a slot descriptions enhanced generative approach for ze…

Dialogue State TrackingTransfer Learning

AISFG: Abundant Information Slot Filling Generator

2022-07-01 · NAACL 2022 7 · Yang Yan, Junda Ye, Zhongbao Zhang, LiWen Wang

As an essential component of task-oriented dialogue systems, slot filling requires enormous labeled training data in a certain domain. However, in most cases, there is little or no target domain training data is availabl…

Few-Shot Learningslot-fillingSlot FillingTask-Oriented Dialogue Systems

QA-Driven Zero-shot Slot Filling with Weak Supervision Pretraining

2021-08-01 · ACL 2021 5 · Xinya Du, Luheng He, Qi Li, Dian Yu 외

Slot-filling is an essential component for building task-oriented dialog systems. In this work, we focus on the zero-shot slot-filling problem, where the model needs to predict slots and their values, given utterances fr…

slot-fillingSlot FillingZero-shot Slot Filling

Generative Zero-Shot Prompt Learning for Cross-Domain Slot Filling with Inverse Prompting

2023-07-06 · Xuefeng Li, LiWen Wang, Guanting Dong, Keqing He 외

Zero-shot cross-domain slot filling aims to transfer knowledge from the labeled source domain to the unlabeled target domain. Existing models either encode slot descriptions and examples or design handcrafted question te…

Prompt Learningslot-fillingSlot Filling

Robust Zero-Shot Cross-Domain Slot Filling with Example Values

2019-06-17 · ACL 2019 7 · Darsh J Shah, Raghav Gupta, Amir A Fayazi, Dilek Hakkani-Tur

Task-oriented dialog systems increasingly rely on deep learning-based slot filling models, usually needing extensive labeled training data for target domains. Often, however, little to no target domain training data may …

slot-fillingSlot FillingZero-shot Slot Filling