@inproceedings{phat-2026-phatthachdau,
    title = "Phatthachdau at {S}em{E}val-2026 Task 9: A Multi-Stage Augment-Judge-Train Pipeline for Multilingual Online Polarization Detection",
    author = "Phat, Phan",
    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-1.208/",
    pages = "1616--1620",
    ISBN = "979-8-89176-414-9",
    abstract = "Address the extreme label imbalance in the Hausa dataset where only 11{\%} of instances are polarized{---}through the Augment-Judge-Train (AJT) pipeline. By leveraging Gemini 2.0 for taxonomy-driven data generation and an LLM-as-a-Judge layer for quality control, we expanded the minority class sixfold. Our ensemble architecture, combining specialized Encoders with LLM-LORA, achieved 1st Place in Hausa (0.8336 Macro-F1) and ranked in the Top 10 for English. These results demonstrate the efficacy of culture-aware synthetic data in enhancing social NLP for low-resource languages."
}