@inproceedings{hofenbitzer-etal-2025-germediq,
    title = "{G}er{M}ed{IQ}: A Resource for Simulated and Synthesized Anamnesis Interview Responses in {G}erman",
    author = {Hofenbitzer, Justin  and
      Sch{\"o}ning, Sebastian  and
      Sebastian, Belle  and
      Lammert, Jacqueline  and
      Modersohn, Luise  and
      Boeker, Martin  and
      Frassinelli, Diego},
    editor = "Zhao, Jin  and
      Wang, Mingyang  and
      Liu, Zhu",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-srw.84/",
    pages = "1064--1078",
    ISBN = "979-8-89176-254-1",
    abstract = "Due to strict privacy regulations, text corpora in non-English clinical contexts are scarce. Consequently, synthetic data generation using Large Language Models (LLMs) emerges as a promising strategy to address this data gap. To evaluate the ability of LLMs in generating synthetic data, we applied them to our novel German Medical Interview Questions Corpus (GerMedIQ), which consists of 4,524 unique, simulated question-response pairs in German. We augmented our corpus by prompting 18 different LLMs to generate responses to the same questions. Structural and semantic evaluations of the generated responses revealed that large-sized language models produced responses comparable to those provided by humans. Additionally, an LLM-as-a-judge study, combined with a human baseline experiment assessing response acceptability, demonstrated that human raters preferred the responses generated by Mistral (124B) over those produced by humans. Nonetheless, our findings indicate that using LLMs for data augmentation in non-English clinical contexts requires caution."
}