@inproceedings{dash-etal-2026-semantic,
    title = "Semantic Vectors at {S}em{E}val-2026 Task 9: Robust Multilingual Polarization Detection via Dual-Encoder Fusion and Expert Ensembling",
    author = "Dash, Ankit  and
      Mittal, Priyanshu  and
      Prashant, Piyush  and
      Saumya, Sunil",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.239/",
    pages = "1903--1910",
    ISBN = "979-8-89176-414-9",
    abstract = "We present SEMANTIC VECTORS, our system for POLAR@SemEval-2026 Task 9 on multilingual online polarization detection across 22 typologically diverse languages. Polarization is frequently conveyed through implicit rhetorical framing, making cross-lingual detection highly challenging. We address this with a Siamese dual-encoder jointly fine-tuning mDeBERTa-v3-base and XLM-ROBERTa-large via 4-bit QLoRA, fused with language-specific expert models (GBERT, Italian BERT, Swahili BERT) through an XGBoost meta-stacker with per-language Platt calibration. Rather than addressing class imbalance, focal loss functions as a hard-example miner, concentrating gradients on subtly framed instances rather than lexically obvious ones. Combined with per-language threshold optimization, our system achieves macro-F1=0.797 and accuracy=0.827 across all 22 languages."
}