GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling

Libo Qin; Fuxuan Wei; Tianbao Xie; Xu Xiao; Wanxiang Che; Ting Liu

doi:10.18653/v1/2021.acl-long.15

January 1, 2021Open Access

GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling

View Full Paper

Ask AI

Bookmark

Share

Key Points

To develop a fast and accurate non-autoregressive architecture that jointly predicts multiple intents and extracts slot entities in spoken dialogue systems.
Designed a non-autoregressive neural network architecture integrating global and local graph interaction mechanisms.
Jointly modeled sentence-level multiple intent classification and token-level slot filling concurrently rather than via sequential autoregressive generation.
Evaluated execution latency and joint task performance on standard multi-intent spoken language understanding benchmark datasets.
Achieved significantly reduced inference latency compared to sequential autoregressive models by generating predictions simultaneously.
Improved joint intent detection and slot filling accuracy over previous baselines by effectively modeling inter-task correlations through graph interactions.

Implication

Computational study introduces a fast non-autoregressive framework for spoken language understanding, demonstrating improved speed and accuracy across multi-intent dialogue benchmarks.

Abstract

Libo Qin, Fuxuan Wei, Tianbao Xie, Xiao Xu, Wanxiang Che, Ting Liu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Qin et al. (2021) studied this question.

synapsesocial.com/papers/6a0ea3ad1c5e2d2319f9a62c https://doi.org/10.18653/v1/2021.acl-long.15

View Full Paper

Ask AI

Bookmark

Share

GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling

Key Points

Implication

Abstract

Discussion

Cite This Study

GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

Keep reading

More related papers