<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-hallucination">
    <titleInfo>
        <title>Hallucination as Exploit: Evidence-Carrying Multimodal Agents</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Guijia</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hao</namePart>
        <namePart type="family">Zheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Harry</namePart>
        <namePart type="family">Yang</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>Multimodal agents increasingly choose tool calls from screenshots, documents, and webpages, where a false perceptual claim can turn hallucination from an answer-quality error into an authorization failure. We formalize this failure as hallucination-to-action conversion (H2AC) and propose evidence-carrying multimodal agents (ECA), in which model text may propose actions but cannot authorize them: constrained DOM/OCR/AX verifiers issue typed certificates for action-critical predicates, and a deterministic gate executes only what the certificates support. Across verifier red-teaming, 7,488 replayed authorization traces, and end-to-end and live-browser evaluations, ECA executes no unsafe action, while GPT-5.4 judge baselines admit 79–99% of unsafe actions under the same threat model; benign actions are occasionally deferred to user confirmation (6.4% on live pages) but never hard-blocked. The guarantee is conditional, extending exactly as far as the verifiers, schemas, and gate implementation that form ECA’s trusted base.</abstract>
    <identifier type="citekey">zhang-etal-2026-hallucination</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.871</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.871/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>18413</start>
            <end>18437</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Hallucination as Exploit: Evidence-Carrying Multimodal Agents
%A Zhang, Guijia
%A Zheng, Hao
%A Yang, Harry
%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 zhang-etal-2026-hallucination
%X Multimodal agents increasingly choose tool calls from screenshots, documents, and webpages, where a false perceptual claim can turn hallucination from an answer-quality error into an authorization failure. We formalize this failure as hallucination-to-action conversion (H2AC) and propose evidence-carrying multimodal agents (ECA), in which model text may propose actions but cannot authorize them: constrained DOM/OCR/AX verifiers issue typed certificates for action-critical predicates, and a deterministic gate executes only what the certificates support. Across verifier red-teaming, 7,488 replayed authorization traces, and end-to-end and live-browser evaluations, ECA executes no unsafe action, while GPT-5.4 judge baselines admit 79–99% of unsafe actions under the same threat model; benign actions are occasionally deferred to user confirmation (6.4% on live pages) but never hard-blocked. The guarantee is conditional, extending exactly as far as the verifiers, schemas, and gate implementation that form ECA’s trusted base.
%R 10.18653/2026.findings-emnlp.871
%U https://aclanthology.org/2026.findings-emnlp.871/
%U https://doi.org/10.18653/2026.findings-emnlp.871
%P 18413-18437

Markdown (Informal)

[Hallucination as Exploit: Evidence-Carrying Multimodal Agents](https://aclanthology.org/2026.findings-emnlp.871/) (Zhang et al., Findings 2026)

• Hallucination as Exploit: Evidence-Carrying Multimodal Agents (Zhang et al., Findings 2026)

ACL

• Guijia Zhang, Hao Zheng, and Harry Yang. 2026. Hallucination as Exploit: Evidence-Carrying Multimodal Agents. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 18413–18437, Budapest, Hungary. Association for Computational Linguistics.