Hate-Alert@DravidianLangTech-EACL2021: Ensembling strategies for Transformer-based Offensive language Detection
arXiv:2102.10084
Abstract
Social media often acts as breeding grounds for different forms of offensive content. For low resource languages like Tamil, the situation is more complex due to the poor performance of multilingual or language-specific models and lack of proper benchmark datasets. Based on this shared task, Offensive Language Identification in Dravidian Languages at EACL 2021, we present an exhaustive exploration of different transformer models, We also provide a genetic algorithm technique for ensembling different models. Our ensembled models trained separately for each language secured the first position in Tamil, the second position in Kannada, and the first position in Malayalam sub-tasks. The models and codes are provided.
arXiv comment6 pages, 1 figure, 3 tables, code available at https://github.com/Debjoy10/Hate-Alert-DravidianLangTech
References in corpus (6)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Decoupled Weight Decay Regularization
- HuggingFace's Transformers: State-of-the-art Natural Language Processing
- Offensive Language Identification in Greek
- Offensive Language and Hate Speech Detection for Danish