@inproceedings{inostroza-etal-2026-cat,
    title = "Where the Cat Sat: A Multilingual Framework for Spatial Language Understanding",
    author = "Inostroza, Demian  and
      Vylomova, Ekaterina  and
      Kemp, Charles  and
      Carroll, Mae  and
      Li, Wanchun  and
      Mistica, Meladel",
    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.1633/",
    pages = "35344--35362",
    ISBN = "979-8-89176-390-6",
    abstract = "Spatial language understanding is fundamental to tasks from robot navigation to document analysis, yet current work exhibits biases toward English and prepositional marking. We present a multilingual framework and benchmark decomposing spatial relations into surface elements (figure, ground, predicate, markers) and semantic components (dynamicity, stasis). Evaluating frontier LLMs on Spanish, Basque, and Chinese with text-only input, we find high accuracy on figure and ground identification but persistent gaps in two areas: semantic classification of topological and projective relations, and surface identification of morphological spatial markers{---}Basque case affixes proving most challenging at as low as 15.3{\%}. These results suggest that surface parsing does not entail spatial understanding, and that evaluation must include typologically diverse spatial marking strategies."
}