基于多意图融合框架的联合意图识别和槽填充(A Multi-Intent Fusion Framework for Joint Intent Detection and Slot Filling)
Shangjian Yin (尹商鉴), Peijie Huang (黄沛杰), Dongzhu Liang (梁栋柱), Zhuoqi He (何卓棋), Qianer Li (黎倩尔), Yuhong Xu (徐禹洪)
Abstract
“近年来,多意图口语理解(SLU)已经成为自然语言处理领域的研究热点。当前先进的多意图SLU模型采用图-交互式框架进行联合多意图识别和槽位填充,能够有效地捕捉到词元级槽位填充任务的细粒度意图信息,取得了良好的性能。但是,它忽略了联合作用下的意图所包含的丰富信息,没有充分利用多意图信息对槽填充任务进行指引。为此,本文提出了一种基于多意图融合框架(MIFF)的联合多意图识别和槽填充框架,使得模型能够在准确地识别不同意图的同时,利用意图信息为槽填充任务提供更充分的指引。我们在MixATIS和MixSNIPS两个公共数据集上进行了实验,结果表明,我们的模型在性能和效率方面均超过了当前最先进的方法,同时能够有效从单领域数据集泛化到多领域数据集上。”- Anthology ID:
- 2023.ccl-1.5
- Volume:
- Proceedings of the 22nd Chinese National Conference on Computational Linguistics
- Month:
- August
- Year:
- 2023
- Address:
- Harbin, China
- Editors:
- Maosong Sun, Bing Qin, Xipeng Qiu, Jing Jiang, Xianpei Han
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 54–63
- Language:
- Chinese
- URL:
- https://aclanthology.org/2023.ccl-1.5
- DOI:
- Cite (ACL):
- Shangjian Yin, Peijie Huang, Dongzhu Liang, Zhuoqi He, Qianer Li, and Yuhong Xu. 2023. 基于多意图融合框架的联合意图识别和槽填充(A Multi-Intent Fusion Framework for Joint Intent Detection and Slot Filling). In Proceedings of the 22nd Chinese National Conference on Computational Linguistics, pages 54–63, Harbin, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 基于多意图融合框架的联合意图识别和槽填充(A Multi-Intent Fusion Framework for Joint Intent Detection and Slot Filling) (Yin et al., CCL 2023)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2023.ccl-1.5.pdf