MKJ at SemEval-2026 Task 9: A Comparative Study of Generalist, Specialist, and Ensemble Strategies for Multilingual Polarization

TL;DR

This paper systematically compares generalist, specialist, and ensemble strategies for multilingual polarization detection across 22 languages, proposing a language-adaptive framework that improves performance over standard models.

Contribution

It introduces a language-adaptive approach switching between models based on development performance, enhancing multilingual polarization detection accuracy.

Findings

A final macro F1 score of 0.796 was achieved.

Language-specific specialists outperform generalists on scripts with distinct characters.

Cross-lingual augmentation often underperformed native architecture selection.

Abstract

We present a systematic study of multilingual polarization detection across 22 languages for SemEval-2026 Task 9 (Subtask 1), contrasting multilingual generalists with language-specific specialists and hybrid ensembles. While a standard generalist like XLM-RoBERTa suffices when its tokenizer aligns with the target text, it may struggle with distinct scripts (e.g., Khmer, Odia) where monolingual specialists yield significant gains. Rather than enforcing a single universal architecture, we adopt a language-adaptive framework that switches between multilingual generalists, language-specific specialists, and hybrid ensembles based on development performance. Additionally, cross-lingual augmentation via NLLB-200 yielded mixed results, often underperforming native architecture selection and degrading morphologically rich tracks. Our final system achieves an overall macro-averaged F1 score of 0.796 and an average accuracy of 0.826 across all 22 tracks. Code and final test predictions are publicly available at: https://github.com/Maziarkiani/SemEval2026-Task9-Subtask1-Polarization.