@inproceedings{zeng-etal-2026-cogevolve,
    title = "{C}og{E}volve: A Multimodal Benchmark for Evaluating Relational Reasoning in Semantic Extension",
    author = "Zeng, Jingjie  and
      Li, Huayang  and
      Yang, Liang  and
      Sun, Yuanyuan  and
      Zhang, Shaowu  and
      Lin, Hongfei",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1190/",
    pages = "25943--25960",
    ISBN = "979-8-89176-390-6",
    abstract = "Human cognition excels at extending knowledge through analogy, where word meanings evolve along structured pathways from concrete prototypes to abstract senses via metaphor and metonymy. Do Large Language Models (LLMs) internalize this generative logic, or merely mimic statistical patterns? To investigate this, we introduce CogEvolve, a cognitive linguistic benchmark designed to test these evolutionary pathways across textual and visual modalities. Our evaluation reveals a distinct cognitive profile: models function as ``Super-Associators'' expert at static recognition yet fail at causal reasoning. In text, they exhibit a Frequency-Primacy Conflation, confusing statistical prevalence with cognitive basicness. Crucially, this reasoning collapses further in the visual domain. We term this deficit the Ungrounded Arrow: models possess high-fidelity concept representations (the ``dots'') but lack the transformational operators (the ``arrows'') essential for true relational understanding."
}