DeepSemantics at SemEval-2026 Task 9: Label-Wise Optimization with Adaptive Focal Loss for Polarization Manifestation Identification
Year: 2026
Type: conference-paper
Abstract: In this paper, we present our system for SemEval-2026 Task 9, which focuses on the fine-grained identification of polarization manifestations in multilingual social media content.Our approach combines transformer-based encoders (RoBERTa-base for English and Afro-XLM-R-small for Hausa) within a One-vs-Rest (OvR) framework, complemented by controlled oversampling, Adaptive Focal Loss, and labelwise threshold optimization.To mitigate severe class imbalance and label sparsity, we adopt language-specific optimization strategies supported by pairwise χ 2 independence analysis.Our system achieves macro-F1 scores of 0.464 in English and 0.192 in Hausa on the official test sets, ranking 5 th in Hausa and 14 th in English on the official leaderboard.Our code is publicly available 1 .
Authors: Eliasse Tiao, Josue Edou, Mahugnon Gohouede
Institution: African Institute for Mathematical Sciences
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166876889