2000 character limit reached
SJ_AJ@DravidianLangTech-EACL2021: Task-Adaptive Pre-Training of Multilingual BERT models for Offensive Language Identification (2102.01051v2)
Published 1 Feb 2021 in cs.CL
Abstract: In this paper we present our submission for the EACL 2021-Shared Task on Offensive Language Identification in Dravidian languages. Our final system is an ensemble of mBERT and XLM-RoBERTa models which leverage task-adaptive pre-training of multilingual BERT models with a masked LLMing objective. Our system was ranked 1st for Kannada, 2nd for Malayalam and 3rd for Tamil.