@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://aclanthology.org/2026.findings-acl.1045/",
doi = "10.18653/v1/2026.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."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hashiloni-etal-2026-id10m">
<titleInfo>
<title>ID10M-JAM: Stress-Testing Idiom Identification Under Challenging Context</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="given">Golan</namePart>
<namePart type="family">Hashiloni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lior</namePart>
<namePart type="family">Livyatan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ofri</namePart>
<namePart type="family">Hefetz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alon</namePart>
<namePart type="family">Mannor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bar</namePart>
<namePart type="family">Cohen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kfir</namePart>
<namePart type="family">Bar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<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.</abstract>
<identifier type="citekey">hashiloni-etal-2026-id10m</identifier>
<identifier type="doi">10.18653/v1/2026.findings-acl.1045</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1045/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>20846</start>
<end>20864</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ID10M-JAM: Stress-Testing Idiom Identification Under Challenging Context
%A Hashiloni, Kai Golan
%A Livyatan, Lior
%A Hefetz, Ofri
%A Mannor, Alon
%A Cohen, Bar
%A Bar, Kfir
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F hashiloni-etal-2026-id10m
%X 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.
%R 10.18653/v1/2026.findings-acl.1045
%U https://aclanthology.org/2026.findings-acl.1045/
%U https://doi.org/10.18653/v1/2026.findings-acl.1045
%P 20846-20864
Markdown (Informal)
[ID10M-JAM: Stress-Testing Idiom Identification Under Challenging Context](https://aclanthology.org/2026.findings-acl.1045/) (Hashiloni et al., Findings 2026)
ACL