mlsanityData to 14 Oct 2026

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

arXiv 2604.21370 ↗First posted 23 Apr 2026cs.CL cs.CYcited by 0

Not on the map. Why?

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.

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

21 of 28 references.

References · 21

  1. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  2. 2021DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharingcited by 1,254
  3. 2020AraBERT: Transformer-based Model for Arabic Language Understandingcited by 362
  4. 2020BERTweet: A pre-trained language model for English Tweetscited by 346
  5. 2020How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Modelscited by 318
  6. 2021Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotationscited by 314
  7. 2020Revisiting Pre-Trained Models for Chinese Natural Language Processingcited by 266
  8. 2021MuRIL: Multilingual Representations for Indian Languagescited by 179
  9. 2020ARBERT & MARBERT: Deep Bidirectional Transformers for Arabiccited by 166
  10. 2022Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuningcited by 149
  11. 2020German's Next Language Modelcited by 144
  12. 2021BanglaBERT: Language Model Pretraining and Benchmarks for Low-Resource Language Understanding Evaluation in Banglacited by 125
9 more
  1. 2019Adaptation of Deep Bidirectional Multilingual Transformers for Russian Languagecited by 109
  2. 2020ParsBERT: Transformer-based Model for Persian Language Understandingcited by 93
  3. 2023Spanish Pre-trained BERT Model and Evaluation Datacited by 81
  4. 2022L3Cube-HindBERT and DevBERT: Pre-Trained BERT Transformer models for Devanagari based Hindi and Marathi Languagescited by 35
  5. 2023A Family of Pretrained Transformer Language Models for Russiancited by 27
  6. 2021Hate and Offensive Speech Detection in Hindi and Marathicited by 18
  7. 2025POLAR: A Benchmark for Multilingual, Multicultural, and Multi-Event Online Polarizationcited by 4
  8. 2024Fine-tuning Transformer-based Encoder for Turkish Language Understanding Taskscited by 3
  9. 2026Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classificationcited by 2

Built on by

Not cited yet.