Exploring Portuguese Hate Speech Detection with Transformers

Social Media platforms, vital for debate and communication, also grapple with misinformation and hateful comments. This work examines the detection of hate speech in Portuguese, contemplating its unique linguistic and cultural nuances. Leveraging Transformer-based models and different training and activation strategies, eight models with variations in architecture, size, and pre-training corpora are evaluated. Our findings show that, even though large generative models with enhanced prompts exhibited promising results, tuned small language models are still superior in addressing this task.