SwiftSLU: An Framework for Cost-Efficient Dataset Construction and Boundary-Modifier-Aware Joint Model in Domain-Specific SLU

  • Dongdong Yang
  • , Chong Feng*
  • , Xinyan Li
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

SLU datasets are highly domain-specific and require a token-level fine-grained annotation, which lead to extremely high cost for manual BIO labeling. In this paper, We introduce a cost-efficient dataset construction approach, demonstrated through the development of VehiCom, a vehicle commands SLU dataset in Chinese built from scratch. We also propose JointVehiCom, a boundary-aware and modifier-aware joint SLU model that excels in accurately identifying entity boundaries in utterances and recognizing entities within auxiliary components, achieving state-of-the-art performance.

Original languageEnglish
Title of host publication10th International Conference on Computer and Communication Systems, ICCCS 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages655-661
Number of pages7
ISBN (Electronic)9798331523145
DOIs
Publication statusPublished - 2025
Externally publishedYes
Event10th International Conference on Computer and Communication Systems, ICCCS 2025 - Chengdu, China
Duration: 18 Apr 202521 Apr 2025

Publication series

Name10th International Conference on Computer and Communication Systems, ICCCS 2025

Conference

Conference10th International Conference on Computer and Communication Systems, ICCCS 2025
Country/TerritoryChina
CityChengdu
Period18/04/2521/04/25

Keywords

  • Intent Detection
  • Slot Filling
  • Spoken Language Understanding
  • Vehicle Command

Fingerprint

Dive into the research topics of 'SwiftSLU: An Framework for Cost-Efficient Dataset Construction and Boundary-Modifier-Aware Joint Model in Domain-Specific SLU'. Together they form a unique fingerprint.

Cite this