@inproceedings{hashiloni-etal-2026-id10m,
    title = "{ID}10{M}-{JAM}: Stress-Testing Idiom Identification Under Challenging Context",
    author = "Hashiloni, Kai Golan  and
      Livyatan, Lior  and
      Hefetz, Ofri  and
      Mannor, Alon  and
      Cohen, Bar  and
      Bar, Kfir",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-form-platform/2026.findings-acl.1045/",
    pages = "20846--20864",
    ISBN = "979-8-89176-395-1",
    abstract = "Large language models (LLMs) achieve strong performance on idiom identification benchmarks, yet their robustness to misleading contextual signals remains largely untested. We introduce ID10M-JAM, an adversarial extension of the ID10M dataset designed to jam model understanding by injecting coherent but conflicting context before each target sentence. For every sentence containing a potential idiomatic expression (PIE), we construct variants that deliberately invert contextual expectations: placing literal cues before idiomatic uses and idiomatic cues before literal ones. All variants are validated by human annotators to ensure naturalness and unambiguous interpretation for human readers. ID10M-JAM exposes systematic vulnerabilities in LLMs' contextual reasoning, pushing idiom identification to its breaking point."
}