<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hosain-etal-2026-hi">
    <titleInfo>
        <title>HI-REASO: A Multi-Level Multi-Faceted Hindi Mathematical Evaluation Suite for Foundation Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Md</namePart>
        <namePart type="given">Tanzib</namePart>
        <namePart type="family">Hosain</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Rajan</namePart>
        <namePart type="given">Das</namePart>
        <namePart type="family">Gupta</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zannatul</namePart>
        <namePart type="given">Fardaush</namePart>
        <namePart type="family">Tripty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Md</namePart>
        <namePart type="given">Kishor</namePart>
        <namePart type="family">Morol</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Large Language Models (LLMs) have shown impressive capabilities across many language understanding tasks; however, their ability to perform numerical reasoning remains fragile, particularly outside English. This limitation is especially concerning for Hindi, a language with hundreds of millions of speakers but relatively little focused evaluation in numerically grounded reasoning. In this work, we present HI-REASO, a benchmark designed to systematically evaluate numerical reasoning in Hindi. The benchmark comprises six complementary task types and 3k instances drawn from educational materials, real-world problem formulations, and commonsense reasoning scenarios. We assess a range of state-of-the-art LLMs using Hindi-native prompting as well as cross-lingual prompting and cross-lingual chain-of-thought reasoning. Across all settings, models consistently underperform compared to human baselines, with the largest performance degradation observed when reasoning is conducted directly in Hindi. While cross-lingual strategies offer partial improvements, they do not close the gap. Our findings reveal fundamental shortcomings in current LLMs’ numerical reasoning abilities under Hindi-native conditions and motivate the development of models and training paradigms that better support linguistically diverse numerical reasoning. We publicly release the code and data at https://github.com/kraritt/hi-reaso.</abstract>
    <identifier type="citekey">hosain-etal-2026-hi</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.481/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>10090</start>
            <end>10112</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T HI-REASO: A Multi-Level Multi-Faceted Hindi Mathematical Evaluation Suite for Foundation Models
%A Hosain, Md Tanzib
%A Gupta, Rajan Das
%A Tripty, Zannatul Fardaush
%A Morol, Md Kishor
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F hosain-etal-2026-hi
%X Large Language Models (LLMs) have shown impressive capabilities across many language understanding tasks; however, their ability to perform numerical reasoning remains fragile, particularly outside English. This limitation is especially concerning for Hindi, a language with hundreds of millions of speakers but relatively little focused evaluation in numerically grounded reasoning. In this work, we present HI-REASO, a benchmark designed to systematically evaluate numerical reasoning in Hindi. The benchmark comprises six complementary task types and 3k instances drawn from educational materials, real-world problem formulations, and commonsense reasoning scenarios. We assess a range of state-of-the-art LLMs using Hindi-native prompting as well as cross-lingual prompting and cross-lingual chain-of-thought reasoning. Across all settings, models consistently underperform compared to human baselines, with the largest performance degradation observed when reasoning is conducted directly in Hindi. While cross-lingual strategies offer partial improvements, they do not close the gap. Our findings reveal fundamental shortcomings in current LLMs’ numerical reasoning abilities under Hindi-native conditions and motivate the development of models and training paradigms that better support linguistically diverse numerical reasoning. We publicly release the code and data at https://github.com/kraritt/hi-reaso.
%U https://aclanthology.org/2026.findings-emnlp.481/
%P 10090-10112

Markdown (Informal)

[HI-REASO: A Multi-Level Multi-Faceted Hindi Mathematical Evaluation Suite for Foundation Models](https://aclanthology.org/2026.findings-emnlp.481/) (Hosain et al., Findings 2026)

• HI-REASO: A Multi-Level Multi-Faceted Hindi Mathematical Evaluation Suite for Foundation Models (Hosain et al., Findings 2026)

ACL

• Md Tanzib Hosain, Rajan Das Gupta, Zannatul Fardaush Tripty, and Md Kishor Morol. 2026. HI-REASO: A Multi-Level Multi-Faceted Hindi Mathematical Evaluation Suite for Foundation Models. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 10090–10112, Budapest, Hungary. Association for Computational Linguistics.