Search Results for author: Blaz Skrlj

Found 1 papers, 0 papers with code

To BAN or not to BAN: Bayesian Attention Networks for Reliable Hate Speech Detection

no code implementations10 Jul 2020 Kristian Miok, Blaz Skrlj, Daniela Zaharie, Marko Robnik-Sikonja

Our approach not only improves the classification performance of the state-of-the-art multilingual BERT model but the computed reliability scores also significantly reduce the workload in an inspection of ofending cases and reannotation campaigns.

Classification General Classification +2

Cannot find the paper you are looking for? You can Submit a new open access paper.