Improved Word Sense Disambiguation with Enhanced Sense Representations

Yang Song, Xin Cai Ong, Hwee Tou Ng, Qian Lin


Abstract
Current state-of-the-art supervised word sense disambiguation (WSD) systems (such as GlossBERT and bi-encoder model) yield surprisingly good results by purely leveraging pre-trained language models and short dictionary definitions (or glosses) of the different word senses. While concise and intuitive, the sense gloss is just one of many ways to provide information about word senses. In this paper, we focus on enhancing the sense representations via incorporating synonyms, example phrases or sentences showing usage of word senses, and sense gloss of hypernyms. We show that incorporating such additional information boosts the performance on WSD. With the proposed enhancements, our system achieves an F1 score of 82.0% on the standard benchmark test dataset of the English all-words WSD task, surpassing all previous published scores on this benchmark dataset.
Anthology ID:
2021.findings-emnlp.365
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
4311–4320
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.365
DOI:
10.18653/v1/2021.findings-emnlp.365
Bibkey:
Cite (ACL):
Yang Song, Xin Cai Ong, Hwee Tou Ng, and Qian Lin. 2021. Improved Word Sense Disambiguation with Enhanced Sense Representations. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 4311–4320, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Improved Word Sense Disambiguation with Enhanced Sense Representations (Song et al., Findings 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.findings-emnlp.365.pdf
Video:
 https://aclanthology.org/2021.findings-emnlp.365.mp4
Code
 nusnlp/esr
Data
Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison