@inproceedings{rabbani-etal-2025-hoax,
    title = "Hoax Terminators@{LT}-{EDI} 2025: {C}har{BERT}{'}s dominance over {LLM} Models in the Detection of Racial Hoaxes in Code-Mixed {H}indi-{E}nglish Social Media Data",
    author = "Rabbani, Abrar Hafiz  and
      Droba, Diganta Das  and
      Labib, Momtazul Arefin  and
      Rahman, Samia  and
      Murad, Hasan",
    editor = "Gkirtzou, Katerina  and
      {\v{Z}}itnik, Slavko  and
      Gracia, Jorge  and
      Gromann, Dagmar  and
      di Buono, Maria Pia  and
      Monti, Johanna  and
      Ionov, Maxim",
    booktitle = "Proceedings of the 5th Conference on Language, Data and Knowledge: Fifth Workshop on Language Technology for Equality, Diversity, Inclusion",
    month = sep,
    year = "2025",
    address = "Naples, Italy",
    publisher = "Unior Press",
    url = "https://preview.aclanthology.org/corrections-2025-10/2025.ltedi-1.27/",
    pages = "160--171",
    ISBN = "978-88-6719-334-9",
    abstract = "This paper presents our system for the LT-EDI 2025 Shared Task on Racial Hoax Detection, addressing the critical challenge of identifying racially charged misinformation in code-mixed Hindi-English (Hinglish) social media{---}a low-resource, linguistically complex domain with real-world impact. We adopt a two-pronged strategy, independently fine-tuning a transformer-based model and a large language model. CharBERT was optimized using Optuna, while XLM-RoBERTa and DistilBERT were fine-tuned for the classification task. FLAN-T5-base was fine-tuned with SMOTE-based oversampling, semantic-preserving back translation, and prompt engineering, whereas LLaMA was used solely for inference. Our preprocessing included Hinglish-specific normalization, noise reduction, sentiment-aware corrections and a custom weighted loss to emphasize the minority Hoax class. Despite using FLAN-T5-base due to resource limits, our models performed well. CharBERT achieved a macro F1 of 0.70 and FLAN-T5 followed at 0.69, both outperforming baselines like DistilBERT and LLaMA-3.2-1B. Our submission ranked 4th of 11 teams, underscoring the promise of our approach for scalable misinformation detection in code-switched contexts. Future work will explore larger LLMs, adversarial training and context-aware decoding."
}