@inproceedings{garcia-etal-2026-retrieval,
    title = "Retrieval-Augmented Generation for Clinical Question Answering in {P}ortuguese Drug Leaflets: Benefits and Limitations",
    author = "Garcia, Gabriel Lino  and
      Paiola, Pedro Henrique  and
      Correia, Jo{\~a}o Vitor Mariano  and
      Rodrigues, Douglas  and
      Papa, Jo{\~a}o Paulo",
    editor = "Souza, Marlo  and
      de-Dios-Flores, Iria  and
      Santos, Diana  and
      Freitas, Larissa  and
      Souza, Jackson Wilke da Cruz  and
      Ribeiro, Eug{\'e}nio",
    booktitle = "Proceedings of the 17th International Conference on Computational Processing of {P}ortuguese ({PROPOR} 2026) - Vol. 2",
    month = apr,
    year = "2026",
    address = "Salvador, Brazil",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-dnd/2026.propor-2.18/",
    pages = "112--120",
    ISBN = "979-8-89176-387-6",
    abstract = "Retrieval-Augmented Generation (RAG) is proposed to reduce hallucination and improve grounding in clinical language models, yet its effectiveness across different levels of clinical reasoning remains unclear. We conducted a controlled evaluation of medication-related question answering in Portuguese using over 7,000 Brazilian regulatory drug leaflets and a complementary clinical benchmark derived from national medical licensing examinations (Revalida and Fuvest). Retrieval substantially improved factual recall and clinical coherence in medication-specific queries, increasing F1 from 0.276 to 0.412. However, naive retrieval did not consistently improve complex clinical reasoning and sometimes reduced accuracy compared to a parametric-only baseline. We identify retrieval-induced anchoring bias, where partially relevant evidence shifts model decisions toward clinically incorrect conclusions. Critique-based and adaptive retrieval mitigated this effect and achieved the highest clinical benchmark accuracy (54.25{\%}). Clinically grounded evaluation dimensions revealed safety-relevant differences beyond traditional NLP metrics. These results show that retrieval augmentation is effective in regulatory settings but requires adaptive control for higher-level clinical reasoning."
}