<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="p-m-etal-2026-riddle">
    <titleInfo>
        <title>The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs Using Indian Riddles</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Abhinav</namePart>
        <namePart type="family">P M</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ojasva</namePart>
        <namePart type="family">Saxena</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Oswald</namePart>
        <namePart type="family">C</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Parameswari</namePart>
        <namePart type="family">Krishnamurthy</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Stelios</namePart>
            <namePart type="family">Piperidis</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Núria</namePart>
            <namePart type="family">Bel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Henk</namePart>
            <namePart type="family">van den Heuvel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nancy</namePart>
            <namePart type="family">Ide</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Simon</namePart>
            <namePart type="family">Krek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonio</namePart>
            <namePart type="family">Toral</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resource Association</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The extent to which large language models (LLMs) can perform culturally grounded reasoning across non-English languages remains underexplored. This paper examines the reasoning and self-assessment abilities of LLMs across seven major Indian languages- Bengali, Gujarati, Hindi, Kannada, Malayalam, Tamil, and Telugu. We introduce a multilingual riddle dataset combining traditional riddles with context-reconstructed variants and evaluate five LLMs- Gemini 2.5 Pro, Gemini 2.5 Flash, Mistral-Saba, LLaMA-4-Scout, and LLaMA-4-Maverick under seven prompting strategies. In the first stage, we assess riddle-solving performance and find that while Gemini 2.5 Pro performs best overall, few-shot methods yield only marginal gains, and accuracy varies notably across languages. In the second stage, we conduct a self-evaluation experiment to measure reasoning consistency. The results reveal a key finding: a model’s initial accuracy is inversely correlated with its ability to identify its own mistakes. Top-performing models such as Gemini 2.5 Pro are overconfident (4.34% True Negative Rate), whereas lower-performing models like LLaMA-4-Scout are substantially more self-aware (42.09% True Negative Rate). These results point to clear gaps in multilingual reasoning and highlight the need for models that not only reason effectively but also recognize their own limitations.</abstract>
    <identifier type="citekey">p-m-etal-2026-riddle</identifier>
    <identifier type="doi">10.63317/2pgfbjkdofoe</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.430/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>5516</start>
            <end>5527</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs Using Indian Riddles
%A P M, Abhinav
%A Saxena, Ojasva
%A C, Oswald
%A Krishnamurthy, Parameswari
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F p-m-etal-2026-riddle
%X The extent to which large language models (LLMs) can perform culturally grounded reasoning across non-English languages remains underexplored. This paper examines the reasoning and self-assessment abilities of LLMs across seven major Indian languages- Bengali, Gujarati, Hindi, Kannada, Malayalam, Tamil, and Telugu. We introduce a multilingual riddle dataset combining traditional riddles with context-reconstructed variants and evaluate five LLMs- Gemini 2.5 Pro, Gemini 2.5 Flash, Mistral-Saba, LLaMA-4-Scout, and LLaMA-4-Maverick under seven prompting strategies. In the first stage, we assess riddle-solving performance and find that while Gemini 2.5 Pro performs best overall, few-shot methods yield only marginal gains, and accuracy varies notably across languages. In the second stage, we conduct a self-evaluation experiment to measure reasoning consistency. The results reveal a key finding: a model’s initial accuracy is inversely correlated with its ability to identify its own mistakes. Top-performing models such as Gemini 2.5 Pro are overconfident (4.34% True Negative Rate), whereas lower-performing models like LLaMA-4-Scout are substantially more self-aware (42.09% True Negative Rate). These results point to clear gaps in multilingual reasoning and highlight the need for models that not only reason effectively but also recognize their own limitations.
%R 10.63317/2pgfbjkdofoe
%U https://aclanthology.org/2026.lrec-1.430/
%U https://doi.org/10.63317/2pgfbjkdofoe
%P 5516-5527

Markdown (Informal)

[The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs Using Indian Riddles](https://aclanthology.org/2026.lrec-1.430/) (P M et al., LREC 2026)

• The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs Using Indian Riddles (P M et al., LREC 2026)

ACL

• Abhinav P M, Ojasva Saxena, Oswald C, and Parameswari Krishnamurthy. 2026. The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs Using Indian Riddles. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 5516–5527, Palma de Mallorca, Spain. ELRA Language Resource Association.