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

arXiv: https://arxiv.org/abs/2604.21370 · HTML: https://mlsanity.com/paper/2604.21370/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 28 references that are in the map.
- First posted on arXiv 23 Apr 2026.

## Builds on (most cited first, 21 of 21)

- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [BERTweet: A pre-trained language model for English Tweets](https://mlsanity.com/paper/2005.10200/) (2020)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Revisiting Pre-Trained Models for Chinese Natural Language Processing](https://mlsanity.com/paper/2004.13922/) (2020)
- [MuRIL: Multilingual Representations for Indian Languages](https://mlsanity.com/paper/2103.10730/) (2021)
- [ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic](https://mlsanity.com/paper/2101.01785/) (2020)
- [Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning](https://mlsanity.com/paper/2204.06487/) (2022)
- [German's Next Language Model](https://mlsanity.com/paper/2010.10906/) (2020)
- [BanglaBERT: Language Model Pretraining and Benchmarks for Low-Resource Language Understanding Evaluation in Bangla](https://mlsanity.com/paper/2101.00204/) (2021)
- [Adaptation of Deep Bidirectional Multilingual Transformers for Russian Language](https://mlsanity.com/paper/1905.07213/) (2019)
- [ParsBERT: Transformer-based Model for Persian Language Understanding](https://mlsanity.com/paper/2005.12515/) (2020)
- [Spanish Pre-trained BERT Model and Evaluation Data](https://mlsanity.com/paper/2308.02976/) (2023)
- [L3Cube-HindBERT and DevBERT: Pre-Trained BERT Transformer models for Devanagari based Hindi and Marathi Languages](https://mlsanity.com/paper/2211.11418/) (2022)
- [A Family of Pretrained Transformer Language Models for Russian](https://mlsanity.com/paper/2309.10931/) (2023)
- [Hate and Offensive Speech Detection in Hindi and Marathi](https://mlsanity.com/paper/2110.12200/) (2021)
- [POLAR: A Benchmark for Multilingual, Multicultural, and Multi-Event Online Polarization](https://mlsanity.com/paper/2505.20624/) (2025)
- [Fine-tuning Transformer-based Encoder for Turkish Language Understanding Tasks](https://mlsanity.com/paper/2401.17396/) (2024)
- [Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classification](https://mlsanity.com/paper/2601.16278/) (2026)

## Abstract (arXiv)

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.