Abstract
In this paper, we advocate for using large pre-trained monolingual language models in cross lingual zero-shot word sense disambiguation (WSD) coupled with a contextualized mapping mechanism. We also report rigorous experiments that illustrate the effectiveness of employing sparse contextualized word representations obtained via a dictionary learning procedure. Our experimental results demonstrate that the above modifications yield a significant improvement of nearly 6.5 points of increase in the average F-score (from 62.0 to 68.5) over a collection of 17 typologically diverse set of target languages. We release our source code for replicating our experiments at https://github.com/begab/sparsity_makes_sense.- Anthology ID:
- 2022.naacl-main.176
- Volume:
- Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
- Month:
- July
- Year:
- 2022
- Address:
- Seattle, United States
- Venue:
- NAACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2459–2471
- Language:
- URL:
- https://aclanthology.org/2022.naacl-main.176
- DOI:
- 10.18653/v1/2022.naacl-main.176
- Cite (ACL):
- Gábor Berend. 2022. Combating the Curse of Multilinguality in Cross-Lingual WSD by Aligning Sparse Contextualized Word Representations. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2459–2471, Seattle, United States. Association for Computational Linguistics.
- Cite (Informal):
- Combating the Curse of Multilinguality in Cross-Lingual WSD by Aligning Sparse Contextualized Word Representations (Berend, NAACL 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.naacl-main.176.pdf
- Code
- begab/sparsity_makes_sense
- Data
- word2word