BERTifying Sinhala - A Comprehensive Analysis of Pre-trained Language Models for Sinhala Text Classification
Vinura Dhananjaya, Piyumal Demotte, Surangika Ranathunga, Sanath Jayasena
Abstract
This research provides the first comprehensive analysis of the performance of pre-trained language models for Sinhala text classification. We test on a set of different Sinhala text classification tasks and our analysis shows that out of the pre-trained multilingual models that include Sinhala (XLM-R, LaBSE, and LASER), XLM-R is the best model by far for Sinhala text classification. We also pre-train two RoBERTa-based monolingual Sinhala models, which are far superior to the existing pre-trained language models for Sinhala. We show that when fine-tuned, these pre-trained language models set a very strong baseline for Sinhala text classification and are robust in situations where labeled data is insufficient for fine-tuning. We further provide a set of recommendations for using pre-trained models for Sinhala text classification. We also introduce new annotated datasets useful for future research in Sinhala text classification and publicly release our pre-trained models.- Anthology ID:
- 2022.lrec-1.803
- Volume:
- Proceedings of the Thirteenth Language Resources and Evaluation Conference
- Month:
- June
- Year:
- 2022
- Address:
- Marseille, France
- Editors:
- Nicoletta Calzolari, Frédéric Béchet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Hélène Mazo, Jan Odijk, Stelios Piperidis
- Venue:
- LREC
- SIG:
- Publisher:
- European Language Resources Association
- Note:
- Pages:
- 7377–7385
- Language:
- URL:
- https://aclanthology.org/2022.lrec-1.803
- DOI:
- Cite (ACL):
- Vinura Dhananjaya, Piyumal Demotte, Surangika Ranathunga, and Sanath Jayasena. 2022. BERTifying Sinhala - A Comprehensive Analysis of Pre-trained Language Models for Sinhala Text Classification. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 7377–7385, Marseille, France. European Language Resources Association.
- Cite (Informal):
- BERTifying Sinhala - A Comprehensive Analysis of Pre-trained Language Models for Sinhala Text Classification (Dhananjaya et al., LREC 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-2/2022.lrec-1.803.pdf
- Code
- nlpcuom/Sinhala-text-classification