@inproceedings{uzunoglu-etal-2026-flaw,
    title = "The Flaw of Averages: Measuring Benchmark-Level Distributional Robustness",
    author = "Uzunoglu, Arda  and
      Li, Tianjian  and
      Khashabi, Daniel",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.emnlp-main.1895/",
    pages = "43186--43226",
    ISBN = "979-8-89176-459-0"
}