@inproceedings{momente-etal-2025-triangulating,
    title = "Triangulating {LLM} Progress through Benchmarks, Games, and Cognitive Tests",
    author = "Moment{\`e}, Filippo  and
      Suglia, Alessandro  and
      Giulianelli, Mario  and
      Ferrari, Ambra  and
      Koller, Alexander  and
      Lemon, Oliver  and
      Schlangen, David  and
      Fern{\'a}ndez, Raquel  and
      Bernardi, Raffaella",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.1092/",
    doi = "10.18653/v1/2025.findings-emnlp.1092",
    pages = "20051--20072",
    ISBN = "979-8-89176-335-7",
    abstract = "We examine three evaluation paradigms: standard benchmarks (e.g., MMLU and BBH), interactive games (e.g., Signalling Games or Taboo), and cognitive tests (e.g., for working memory or theory of mind). First, we investigate which of the former two{---}benchmarks or games{---}is most effective at discriminating LLMs of varying quality. Then, inspired by human cognitive assessments, we compile a suite of targeted tests that measure cognitive abilities deemed essential for effective language use, and we investigate their correlation with model performance in benchmarks and games. Our analyses reveal that interactive games are superior to standard benchmarks in discriminating models. Causal and logical reasoning correlate with both static and interactive tests, while differences emerge regarding core executive functions and social/emotional skills, which correlate more with games. We advocate for the development of new interactive benchmarks and targeted cognitive tasks inspired by assessing human abilities but designed specifically for LLMs."
}