@inproceedings{ingimundarson-etal-2026-benchmarks,
    title = "Who Benchmarks the Benchmarks? A Case Study of {LLM} Evaluation in {I}celandic",
    author = "Ingimundarson, Finnur {\'A}g{\'u}st  and
      Fridriksdottir, Steinunn Rut  and
      {\'A}rmannsson, Bjarki  and
      Nowenstein, Iris  and
      Steingr{\'i}msson, Stein{\th}{\'o}r",
    editor = "Piperidis, Stelios  and
      Bel, N{\'u}ria  and
      van den Heuvel, Henk  and
      Ide, Nancy  and
      Krek, Simon  and
      Toral, Antonio",
    booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
    month = may,
    year = "2026",
    address = "Palma de Mallorca, Spain",
    publisher = "ELRA Language Resource Association",
    url = "https://aclanthology.org/2026.lrec-1.369/",
    doi = "10.63317/5nxcp3zw7vdz",
    pages = "4702--4715",
    abstract = "This paper evaluates current Large Language Model (LLM) benchmarking for Icelandic, identifies problems, and calls for improved evaluation methods in low/medium-resource languages in particular. We show that benchmarks that include synthetic or machine-translated data that have not been verified in any way, commonly contain severely flawed test examples that are likely to skew the results and undermine the tests' validity. We warn against the use of such methods without verification in low/medium-resource settings as the translation quality can, at best, only be as good as MT quality for a given language at any given time. Indeed, the results of our quantitative error analysis on existing benchmarks for Icelandic show clear differences between human-authored/-translated benchmarks vs. synthetic or machine-translated benchmarks."
}