Contrastive Bootstrapping for Label Refinement

Shudi Hou, Yu Xia, Muhao Chen, Sujian Li


Abstract
Traditional text classification typically categorizes texts into pre-defined coarse-grained classes, from which the produced models cannot handle the real-world scenario where finer categories emerge periodically for accurate services. In this work, we investigate the setting where fine-grained classification is done only using the annotation of coarse-grained categories and the coarse-to-fine mapping. We propose a lightweight contrastive clustering-based bootstrapping method to iteratively refine the labels of passages. During clustering, it pulls away negative passage-prototype pairs under the guidance of the mapping from both global and local perspectives. Experiments on NYT and 20News show that our method outperforms the state-of-the-art methods by a large margin.
Anthology ID:
2023.acl-short.84
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
976–985
Language:
URL:
https://aclanthology.org/2023.acl-short.84
DOI:
10.18653/v1/2023.acl-short.84
Bibkey:
Cite (ACL):
Shudi Hou, Yu Xia, Muhao Chen, and Sujian Li. 2023. Contrastive Bootstrapping for Label Refinement. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 976–985, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Contrastive Bootstrapping for Label Refinement (Hou et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-short.84.pdf
Video:
 https://aclanthology.org/2023.acl-short.84.mp4