<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="van-dijk-2026-detecting">
    <titleInfo>
        <title>Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Gijs</namePart>
        <namePart type="family">van Dijk</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>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Santosh</namePart>
            <namePart type="family">T.Y.S.S.</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="given">Diego</namePart>
            <namePart type="family">Rodriguez</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ona</namePart>
            <namePart type="family">de Gibert</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">979-8-89176-393-7</identifier>
    </relatedItem>
    <abstract>We propose a lightweight and single-pass uncertainty quantification method for detecting hallucinations in Large Language Models. The method uses attention matrices to estimate uncertainty without requiring repeated sampling or external models. Specifically, we measure the Kullback–Leibler divergence between each attention head’s distribution and a uniform reference distribution, and use these features in a logistic regression probe. Across multiple datasets, task types, and model families, attention divergence is strongly predictive of answer correctness and performs competitively with existing uncertainty estimation methods. We find that this signal is concentrated in middle layers and on factual tokens such as named entities and numbers, suggesting that attention dynamics provides an efficient and interpretable white-box signal of model uncertainty.</abstract>
    <identifier type="citekey">van-dijk-2026-detecting</identifier>
    <location>
        <url>https://aclanthology.org/2026.acl-srw.6/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>60</start>
            <end>75</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals
%A van Dijk, Gijs
%Y T.Y.S.S., Santosh
%Y Rodriguez, Juan Diego
%Y de Gibert, Ona
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-393-7
%F van-dijk-2026-detecting
%X We propose a lightweight and single-pass uncertainty quantification method for detecting hallucinations in Large Language Models. The method uses attention matrices to estimate uncertainty without requiring repeated sampling or external models. Specifically, we measure the Kullback–Leibler divergence between each attention head’s distribution and a uniform reference distribution, and use these features in a logistic regression probe. Across multiple datasets, task types, and model families, attention divergence is strongly predictive of answer correctness and performs competitively with existing uncertainty estimation methods. We find that this signal is concentrated in middle layers and on factual tokens such as named entities and numbers, suggesting that attention dynamics provides an efficient and interpretable white-box signal of model uncertainty.
%U https://aclanthology.org/2026.acl-srw.6/
%P 60-75

Markdown (Informal)

[Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals](https://aclanthology.org/2026.acl-srw.6/) (van Dijk, ACL 2026)

• Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals (van Dijk, ACL 2026)

ACL

• Gijs van Dijk. 2026. Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), pages 60–75, San Diego, California, United States. Association for Computational Linguistics.