Stanceosaurus: Classifying Stance Towards Multilingual Misinformation
This work addresses the need for multilingual stance classification datasets to combat misinformation across diverse cultures, though it is incremental as it builds on existing stance detection methods.
The authors tackled the problem of classifying stance towards misinformation by introducing Stanceosaurus, a large multilingual corpus of 28,033 tweets annotated with stance towards 251 claims, and demonstrated that fine-tuned transformer models achieve cross-lingual generalization with F1 scores of 53.1 on Hindi and 50.4 on Arabic without target-language fine-tuning.
We present Stanceosaurus, a new corpus of 28,033 tweets in English, Hindi, and Arabic annotated with stance towards 251 misinformation claims. As far as we are aware, it is the largest corpus annotated with stance towards misinformation claims. The claims in Stanceosaurus originate from 15 fact-checking sources that cover diverse geographical regions and cultures. Unlike existing stance datasets, we introduce a more fine-grained 5-class labeling strategy with additional subcategories to distinguish implicit stance. Pre-trained transformer-based stance classifiers that are fine-tuned on our corpus show good generalization on unseen claims and regional claims from countries outside the training data. Cross-lingual experiments demonstrate Stanceosaurus' capability of training multi-lingual models, achieving 53.1 F1 on Hindi and 50.4 F1 on Arabic without any target-language fine-tuning. Finally, we show how a domain adaptation method can be used to improve performance on Stanceosaurus using additional RumourEval-2019 data. We make Stanceosaurus publicly available to the research community and hope it will encourage further work on misinformation identification across languages and cultures.