Abstract
Fine-grained Entity Typing (FET) has made great progress based on distant supervision but still suffers from label noise. Existing FET noise learning methods rely on prediction distributions in an instance-independent manner, which causes the problem of confirmation bias. In this work, we propose a clustering-based loss correction framework named Feature Cluster Loss Correction (FCLC), to address these two problems. FCLC first train a coarse backbone model as a feature extractor and noise estimator. Loss correction is then applied to each feature cluster, learning directly from the noisy labels. Experimental results on three public datasets show that FCLC achieves the best performance over existing competitive systems. Auxiliary experiments further demonstrate that FCLC is stable to hyperparameters and it does help mitigate confirmation bias. We also find that in the extreme case of no clean data, the FCLC framework still achieves competitive performance.- Anthology ID:
- 2022.acl-long.141
- Volume:
- Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Month:
- May
- Year:
- 2022
- Address:
- Dublin, Ireland
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1997–2006
- Language:
- URL:
- https://aclanthology.org/2022.acl-long.141
- DOI:
- 10.18653/v1/2022.acl-long.141
- Cite (ACL):
- Kunyuan Pang, Haoyu Zhang, Jie Zhou, and Ting Wang. 2022. Divide and Denoise: Learning from Noisy Labels in Fine-Grained Entity Typing with Cluster-Wise Loss Correction. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1997–2006, Dublin, Ireland. Association for Computational Linguistics.
- Cite (Informal):
- Divide and Denoise: Learning from Noisy Labels in Fine-Grained Entity Typing with Cluster-Wise Loss Correction (Pang et al., ACL 2022)
- PDF:
- https://preview.aclanthology.org/nodalida-main-page/2022.acl-long.141.pdf