Abstract
BERT has been shown to be extremely effective on a wide variety of natural language processing tasks, including sentiment analysis and emotion detection. However, the proposed pretraining objectives of BERT do not induce any sentiment or emotion-specific biases into the model. In this paper, we present Emotion Masked Language Modelling, a variation of Masked Language Modelling aimed at improving the BERT language representation model for emotion detection and sentiment analysis tasks. Using the same pre-training corpora as the original model, Wikipedia and BookCorpus, our BERT variation manages to improve the downstream performance on 4 tasks from emotion detection and sentiment analysis by an average of 1.2% F-1. Moreover, our approach shows an increased performance in our task-specific robustness tests.- Anthology ID:
- 2021.acl-short.38
- Volume:
- Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
- Month:
- August
- Year:
- 2021
- Address:
- Online
- Editors:
- Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
- Venues:
- ACL | IJCNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 286–293
- Language:
- URL:
- https://aclanthology.org/2021.acl-short.38
- DOI:
- 10.18653/v1/2021.acl-short.38
- Cite (ACL):
- Tiberiu Sosea and Cornelia Caragea. 2021. eMLM: A New Pre-training Objective for Emotion Related Tasks. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 286–293, Online. Association for Computational Linguistics.
- Cite (Informal):
- eMLM: A New Pre-training Objective for Emotion Related Tasks (Sosea & Caragea, ACL-IJCNLP 2021)
- PDF:
- https://preview.aclanthology.org/landing_page/2021.acl-short.38.pdf
- Code
- tsosea2/emlm
- Data
- BookCorpus, GoEmotions, SST, SST-2, SST-5