LT@Helsinki at SemEval-2020 Task 12: Multilingual or language-specific BERT?

This paper presents the different models submitted by the LT@Helsinki team\nfor the SemEval 2020 Shared Task 12. Our team participated in sub-tasks A and\nC; titled offensive language identification and offense target identification,\nrespectively. In both cases we used the so-called Bidirectional Encoder\nRepresentation from Transformer (BERT), a model pre-trained by Google and\nfine-tuned by us on the OLID and SOLID datasets. The results show that\noffensive tweet classification is one of several language-based tasks where\nBERT can achieve state-of-the-art results.\n