@inproceedings{kabir-etal-2026-xcr,
title = "{XCR}-Bench: Benchmarking Cross-Cultural Reasoning in {LLM}s via Culture-Specific Items and Hall{'}s Triad",
author = "Kabir, Mohsinul and
Ahmed, Tasnim and
Rahman, Md Mezbaur and
Ji, Shaoxiong and
Alhuzali, Hassan and
Jiang, Yuechen and
Huang, Jimin and
Ananiadou, Sophia",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.741/",
pages = "17179--17209",
ISBN = "979-8-89176-459-0",
abstract = "Cross-cultural competence in large language models (LLMs) requires understanding and adapting Culture-Specific Items (CSIs) across varying cultural contexts. However, progress in evaluating this capability remains limited by the lack of high-quality CSI-annotated corpora with parallel cross-cultural sentence pairs. We introduce XCR-Bench, a Cross$\textbf{(X)}$-$\textbf{C}$ultural $\textbf{R}$easoning $\textbf{Bench}$mark containing $\textbf{4.1k}$ parallel sentences and $\textbf{1{,}098}$ CSIs across three reasoning tasks. XCR-Bench integrates Newmark{'}s CSI framework with Hall{'}s Triad of Culture, enabling evaluation across levels of cultural visibility: from observable practices to implicit social norms and values. Experiments on eight multilingual LLMs show that state-of-the-art models exhibit consistent weaknesses in identifying and adapting specific categories of CSIs, revealing a gap between surface-level recall and explicit cultural reasoning. Performance declines significantly in culturally sensitive categories and at deeper cultural levels ($p<0.005$, 8/8 models). We further find that adaptation quality varies systematically across target cultures and across Bengali regional variants, revealing systematic regional lexical and cultural preference patterns and ethno-religious biases in model outputs, even within a single linguistic setting. We publicly release the corpus and code to support future research on cross-cultural NLP."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kabir-etal-2026-xcr">
<titleInfo>
<title>XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall’s Triad</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mohsinul</namePart>
<namePart type="family">Kabir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tasnim</namePart>
<namePart type="family">Ahmed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="given">Mezbaur</namePart>
<namePart type="family">Rahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaoxiong</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hassan</namePart>
<namePart type="family">Alhuzali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuechen</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jimin</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sophia</namePart>
<namePart type="family">Ananiadou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Cross-cultural competence in large language models (LLMs) requires understanding and adapting Culture-Specific Items (CSIs) across varying cultural contexts. However, progress in evaluating this capability remains limited by the lack of high-quality CSI-annotated corpora with parallel cross-cultural sentence pairs. We introduce XCR-Bench, a Cross(X)-Cultural Reasoning Benchmark containing 4.1k parallel sentences and 1,098 CSIs across three reasoning tasks. XCR-Bench integrates Newmark’s CSI framework with Hall’s Triad of Culture, enabling evaluation across levels of cultural visibility: from observable practices to implicit social norms and values. Experiments on eight multilingual LLMs show that state-of-the-art models exhibit consistent weaknesses in identifying and adapting specific categories of CSIs, revealing a gap between surface-level recall and explicit cultural reasoning. Performance declines significantly in culturally sensitive categories and at deeper cultural levels (p<0.005, 8/8 models). We further find that adaptation quality varies systematically across target cultures and across Bengali regional variants, revealing systematic regional lexical and cultural preference patterns and ethno-religious biases in model outputs, even within a single linguistic setting. We publicly release the corpus and code to support future research on cross-cultural NLP.</abstract>
<identifier type="citekey">kabir-etal-2026-xcr</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.741/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>17179</start>
<end>17209</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall’s Triad
%A Kabir, Mohsinul
%A Ahmed, Tasnim
%A Rahman, Md Mezbaur
%A Ji, Shaoxiong
%A Alhuzali, Hassan
%A Jiang, Yuechen
%A Huang, Jimin
%A Ananiadou, Sophia
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F kabir-etal-2026-xcr
%X Cross-cultural competence in large language models (LLMs) requires understanding and adapting Culture-Specific Items (CSIs) across varying cultural contexts. However, progress in evaluating this capability remains limited by the lack of high-quality CSI-annotated corpora with parallel cross-cultural sentence pairs. We introduce XCR-Bench, a Cross(X)-Cultural Reasoning Benchmark containing 4.1k parallel sentences and 1,098 CSIs across three reasoning tasks. XCR-Bench integrates Newmark’s CSI framework with Hall’s Triad of Culture, enabling evaluation across levels of cultural visibility: from observable practices to implicit social norms and values. Experiments on eight multilingual LLMs show that state-of-the-art models exhibit consistent weaknesses in identifying and adapting specific categories of CSIs, revealing a gap between surface-level recall and explicit cultural reasoning. Performance declines significantly in culturally sensitive categories and at deeper cultural levels (p<0.005, 8/8 models). We further find that adaptation quality varies systematically across target cultures and across Bengali regional variants, revealing systematic regional lexical and cultural preference patterns and ethno-religious biases in model outputs, even within a single linguistic setting. We publicly release the corpus and code to support future research on cross-cultural NLP.
%U https://aclanthology.org/2026.emnlp-main.741/
%P 17179-17209
Markdown (Informal)
[XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall’s Triad](https://aclanthology.org/2026.emnlp-main.741/) (Kabir et al., EMNLP 2026)
ACL
- Mohsinul Kabir, Tasnim Ahmed, Md Mezbaur Rahman, Shaoxiong Ji, Hassan Alhuzali, Yuechen Jiang, Jimin Huang, and Sophia Ananiadou. 2026. XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall’s Triad. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 17179–17209, Budapest, Hungary. Association for Computational Linguistics.