2023 · 17 citations · 22 references
ComputingLinguisticsIntercultural CommunicationNatural Language ProcessingSociolinguisticsHate SpeechLanguage ClassificationLanguage ModelingMachine TranslationMultilingual Natural Language ProcessingHate Speech AnnotationHate Speech CorporaLanguage Processing Models
Increasingly, language models and machine translation are becoming valuable tools to help people communicate with others from diverse cultural backgrounds. However, current language models lack cultural awareness because they are trained on data representing only the culture within the dataset. This presents a problem in the context of hate speech classification, where cultural awareness is especially critical. This study aims to quantify the cultural insensitivity of three monolingual (Korean, English, Arabic) hate speech classifiers by evaluating their performance on translated datasets from the other two languages. Our research has revealed that hate speech classifiers evaluated on datasets from other cultures yield significantly lower F1 scores, up to almost 50%. In addition, they produce considerably higher false negative rates, with a magnitude up to five times greater, demonstrating the extent of the cultural gap. The study highlights the severity of cultural insensitivity of language models in hate speech classification.
22
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Nils Reimers, Iryna Gurevych · 2019 · 10.9K citations · Full text
Abusive Language Detection in Online User Content
Chikashi Nobata, Joel Tetreault, Achint Thomas et al. · 2016 · 1.1K citations
Measuring and Mitigating Unintended Bias in Text Classification
Lucas Dixon, John Li, Jeffrey Sorensen et al. · 2018 · 678 citations · Full text
AraBERT: Transformer-based Model for Arabic Language Understanding
Wissam Antoun, Fady Baly, Hazem Hajj · arXiv (Cornell University) · 2020 · 619 citations · Full text
Computing, Arabic, Computer Science +17
Eun Seo Jo, Timnit Gebru · 2020 · 226 citations · Full text