@inproceedings{oh-etal-2026-tug,
    title = "Tug-of-war between idioms' figurative and literal interpretations in {LLM}s",
    author = "Oh, Soyoung  and
      Huang, Xinting  and
      Pink, Mathis  and
      Hahn, Michael  and
      Demberg, Vera",
    editor = "Demberg, Vera  and
      Inui, Kentaro  and
      Marquez, Llu{\'i}s",
    booktitle = "Proceedings of the 19th Conference of the {E}uropean Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-eacl/2026.eacl-long.135/",
    pages = "2942--2958",
    ISBN = "979-8-89176-380-7",
    abstract = "Idioms present a unique challenge for language models due to their non-compositional figurative interpretations, which often strongly diverge from the idiom{'}s literal interpretation. In this paper, we employ causal tracing to systematically analyze how pretrained causal transformers deal with this ambiguity. We localize three mechanisms: (i) Early sublayers and specific attention heads retrieve figurative interpretation, while suppressing literal interpretation. (ii) When disambiguating context precedes the idiom, the model leverages it from the earliest layer and later layers refine the interpretation if the context conflicts with the retrieved interpretation. (iii) Then, selective, competing pathways carry both interpretations: an intermediate pathway that prioritizes the figurative interpretation and a parallel direct route that favors literal interpretation, ensuring that both readings remain available. Our findings provide mechanistic evidence for idiom comprehension in autoregressive transformers."
}