@inproceedings{kartac-etal-2026-ufal,
    title = "{UFAL}-{CUNI} at {S}em{E}val-2026 Task 11: An Efficient Modular Neuro-symbolic Method for Syllogistic Reasoning",
    author = "Kartac, Ivan  and
      Onderkova, Kristyna  and
      Bronec, Jan  and
      Kasner, Zden{\v{e}}k  and
      Lango, Mateusz  and
      Dusek, Ondrej",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.418/",
    pages = "3363--3376",
    ISBN = "979-8-89176-414-9",
    abstract = "This paper describes our system submitted to SemEval-2026 Task 11: Disentangling Content and Formal Reasoning in Large Language Models. We present an efficient modular neuro-symbolic approach, combining a symbolic prover with small reasoning LLMs (4B parameters). The system consists of an LLM-based parser that translates natural language syllogisms to a first-order logic (FOL) representation, an automated theorem prover, and two optional modules: machine translation for multilingual inputs and a symbolic retrieval component for the identification of relevant premises. The system achieves competitive accuracy and relatively low content effect on most subtasks. Our ablations show that this approach outperforms LLM-based zero-shot baselines in this parameter size range, but also reveal limited multilingual capabilities of small LLMs. Finally, we include a discussion of the task{'}s main ranking metric and analyze its limitations."
}