@inproceedings{oh-etal-2026-finest,
    title = "{FINEST}: Improving {LLM} Responses to Sensitive Topics Through Fine-Grained Evaluation",
    author = "Oh, Juhyun  and
      Lee, Nayeon  and
      Jung, Chani  and
      Jin, Jiho  and
      Myung, Junho  and
      Lee, Jongwon  and
      Song, Taieui  and
      Oh, Alice",
    editor = "Demberg, Vera  and
      Inui, Kentaro  and
      Marquez, Llu{\'i}s",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-eacl/2026.findings-eacl.327/",
    pages = "6207--6226",
    ISBN = "979-8-89176-386-9",
    abstract = "Large Language Models (LLMs) often default to overly cautious and vague responses when handling sensitive topics, sacrificing helpfulness for safety. Existing evaluation frameworks lack systematic methods to identify and address specific weaknesses in responses to sensitive topics, making it difficult to improve both safety and helpfulness simultaneously. To address this, we introduce FINEST, a FINE-grained response evaluation taxonomy for Sensitive Topics, which breaks down helpfulness and harmlessness into errors across three main categories: Content, Logic, and Appropriateness. Experiments on a Korean-sensitive question dataset demonstrate that our score- and error-based improvement pipeline, guided by FINEST, significantly improves the model responses across all three categories, outperforming refinement without guidance. Notably, score-based improvement{---}providing category-specific scores and justifications{---}yields the most significant gains, reducing the error sentence ratio for Appropriateness by up to 33.09{\%}. This work lays the foundation for a more explainable and comprehensive evaluation and improvement of LLM responses to sensitive questions."
}