Sexist Content

Research on sexist content focuses on developing and evaluating automated methods for detecting and classifying sexism in online text and multimedia, aiming to mitigate its harmful effects. Current efforts utilize various deep learning architectures, including transformers (like BERT and RoBERTa) and convolutional neural networks, often incorporating techniques like transfer learning, multi-task learning, and data augmentation to improve performance. This interdisciplinary field, bridging computer science and social sciences, is crucial for creating safer online environments and furthering our understanding of how sexism manifests and propagates in digital spaces. The development of robust and explainable detection systems is a key objective, along with addressing the challenges posed by the subjective nature of sexism and the evolution of online language.