@inproceedings{abdelmalak-2025-abdelmalak,
    title = "Abdelmalak at {P}er{A}ns{S}umm 2025: Leveraging a Domain-Specific {BERT} and {LL}a{MA} for Perspective-Aware Healthcare Answer Summarization",
    author = "Abdelmalak, Abanoub",
    editor = "Ananiadou, Sophia  and
      Demner-Fushman, Dina  and
      Gupta, Deepak  and
      Thompson, Paul",
    booktitle = "Proceedings of the Second Workshop on Patient-Oriented Language Processing (CL4Health)",
    month = may,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.cl4health-1.39/",
    doi = "10.18653/v1/2025.cl4health-1.39",
    pages = "428--436",
    ISBN = "979-8-89176-238-1",
    abstract = "The PerAnsSumm Shared Task - CL4Health@NAACL 2025 aims to enhance healthcare community question-answering (CQA) by summarizing diverse user perspectives. It consists of two tasks: identifying and classifying perspective-specific spans (Task A) and generating structured, perspective-specific summaries from question-answer threads (Task B). The dataset used for this task is the PUMA dataset. For Task A, a COVID-Twitter-BERT model pre-trained on COVID-related text from Twitter was employed, improving the model{'}s understanding of relevant vocabulary and context. For Task B, LLaMA was utilized in a prompt-based fashion. The proposed approach achieved 9th place in Task A and 16th place overall, with the best proportional classification F1-score of 0.74."
}