Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework

Wei, Xiao; Wang, Xiaobao; Zhuang, Ning; Wang, Chenyang; Wang, Longbiao; Dang, Jianwu

doi:10.24963/ijcai.2025/921

Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework

Xiao Wei, Xiaobao Wang, Ning Zhuang, Chenyang Wang, Longbiao Wang, Jianwu Dang

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 8286-8294.

https://doi.org/10.24963/ijcai.2025/921

Keywords

Natural Language Processing: NLP: Natural language semantics
Natural Language Processing: NLP: Text classification
Natural Language Processing: NLP: Dialogue and interactive systems
Machine Learning: ML: Semi-supervised learning

Abstract

Intent detection aims to identify user intents from natural language inputs, where supervised methods rely heavily on labeled in-domain (IND) data and struggle with out-of-domain (OOD) intents, limiting their practical applicability. Generalized Intent Discovery (GID) addresses this by leveraging unlabeled OOD data to discover new intents without additional annotation. However, existing methods focus solely on clustering unsupervised data while neglecting domain adaptation. Therefore, we propose a consistency-driven prototype-prompting framework for GID from the perspective of integrating old and new knowledge, which includes a prototype-prompting framework for transferring old knowledge from external sources, and a hierarchical consistency constraint for learning new knowledge from target domains. We conducted extensive experiments and the results show that our method significantly outperforms all baseline methods, achieving state-of-the-art results, which strongly demonstrates the effectiveness and generalization of our methods. Our source code is publicly available at https://github.com/smileix/cpp.

Xiao Wei, Xiaobao Wang, Ning Zhuang, Chenyang Wang, Longbiao Wang, and Jianwu Dang. 2025. Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 8286–8294. https://doi.org/10.24963/ijcai.2025/921

@inproceedings{ijcai2025p921,
  title     = {Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework},
  author    = {Wei, Xiao and Wang, Xiaobao and Zhuang, Ning and Wang, Chenyang and Wang, Longbiao and Dang, Jianwu},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {8286--8294},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/921},
  url       = {https://doi.org/10.24963/ijcai.2025/921},
}

TY  - CPAPER
TI  - Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework
AU  - Wei, Xiao
AU  - Wang, Xiaobao
AU  - Zhuang, Ning
AU  - Wang, Chenyang
AU  - Wang, Longbiao
AU  - Dang, Jianwu
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 8286
EP  - 8294
DO  - 10.24963/ijcai.2025/921
UR  - https://doi.org/10.24963/ijcai.2025/921
ER  -