MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER

Page view(s)
14
Checked on Apr 16, 2025
MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER
Title:
MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER
Journal Title:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Keywords:
Publication Date:
03 June 2022
Citation:
Zhou, R., Li, X., He, R., Bing, L., Cambria, E., Si, L., & Miao, C. (2022). MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). https://doi.org/10.18653/v1/2022.acl-long.160
Abstract:
Data augmentation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as NER, data augmentation methods often suffer from token-label misalignment, which leads to unsatisfactory performance. In this work, we propose Masked Entity Language Modeling (MELM) as a novel data augmentation framework for low-resource NER. To alleviate the token-label misalignment issue, we explicitly inject NER labels into sentence context, and thus the fine-tuned MELM is able to predict masked entity tokens by explicitly conditioning on their labels. Thereby, MELM generates high-quality augmented data with novel entities, which provides rich entity regularity knowledge and boosts NER performance. When training data from multiple languages are available, we also integrate MELM with codemixing for further improvement. We demonstrate the effectiveness of MELM on monolingual, cross-lingual and multilingual NER across various low-resource levels. Experimental results show that our MELM presents substantial improvement over the baseline methods.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the Agency for Science, Technology and Research (A*STAR) - AME Programmatic Grant
Grant Reference no. : A18A2b0046
Description:
© 2022 Association for Computational Linguistics
ISSN:
1530-9312
Files uploaded:

File Size Format Action
115-wp4-acl-zhou.pdf 1.42 MB PDF Open