YEZE at SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization via Heterogeneous Ensembling
For researchers and practitioners monitoring online polarization in multilingual social media, this provides a practical ensemble approach, but the contribution is incremental.
The authors tackle multilingual online polarization detection across 22 languages with three subtasks. Their heterogeneous ensemble of XLM-RoBERTa-large and mDeBERTa-v3-base with class weighting achieves strong performance, though no specific numbers are reported.
This paper presents our system for SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization, which identifies polarized social media content in 22 languages through three subtasks: binary detection, target classification, and manifestation identification. We propose a heterogeneous ensemble of multilingual pretrained models, combining XLM-RoBERTa-large and mDeBERTa-v3-base. We investigate techniques such as multi-task learning, translation-based data augmentation, and class weighting to improve classification performance under severe label imbalance. Our findings indicate that independent task modeling combined with class weighting is more effective.