<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kim-etal-2026-k-prism">
    <titleInfo>
        <title>K-Prism: A Korean Diagnostic Benchmark for Multi-faceted Hallucination in Vision-Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Min</namePart>
        <namePart type="given">Hyuk</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dong</namePart>
        <namePart type="given">Hyok</namePart>
        <namePart type="family">Shin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jimi</namePart>
        <namePart type="family">Lee</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jin</namePart>
        <namePart type="given">Yea</namePart>
        <namePart type="family">Jang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruifeng</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ayu</namePart>
            <namePart type="family">Purwarianti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tong</namePart>
            <namePart type="family">Rong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jun</namePart>
            <namePart type="family">Suzuki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-440-8</identifier>
    </relatedItem>
    <abstract>Vision-language models (VLMs) often produce hallucinated outputs that are not supported by visual or textual evidence. While existing benchmarks have advanced hallucination evaluation, they primarily target object-level or task-specific errors and often assess different failure types on separate samples. This design limits direct comparison of hallucination patterns under shared evidence, and most prior evaluations remain English-centric. We introduce K-Prism, a Korean diagnostic benchmark for multi-faceted hallucination analysis in VLMs. K-Prism contains 1,002 multiple-choice questions across text and image tracks, organized into shared-evidence topic groups. Within each group, questions are designed to probe three diagnostic dimensions: factual, reasoning, and misleading. K-Prism uses a controlled dual-mode protocol comparing intact inputs with visually perturbed inputs to examine hallucination under reduced visual evidence. Experiments with closed- and open-source VLMs show dimension-specific degradation rather than a uniform accuracy drop. These findings demonstrate that hallucination in Korean VLMs cannot be captured by a single aggregate score and requires fine-grained diagnostic evaluation across evidence conditions and failure dimensions. Our dataset is publicly available at https://huggingface.co/spaces/KETI-NLP/K-Prism_Benchmark.</abstract>
    <identifier type="citekey">kim-etal-2026-k-prism</identifier>
    <location>
        <url>https://aclanthology.org/2026.aacl-short.41/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>632</start>
            <end>644</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T K-Prism: A Korean Diagnostic Benchmark for Multi-faceted Hallucination in Vision-Language Models
%A Kim, Min Hyuk
%A Shin, Dong Hyok
%A Lee, Jimi
%A Jang, Jin Yea
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-440-8
%F kim-etal-2026-k-prism
%X Vision-language models (VLMs) often produce hallucinated outputs that are not supported by visual or textual evidence. While existing benchmarks have advanced hallucination evaluation, they primarily target object-level or task-specific errors and often assess different failure types on separate samples. This design limits direct comparison of hallucination patterns under shared evidence, and most prior evaluations remain English-centric. We introduce K-Prism, a Korean diagnostic benchmark for multi-faceted hallucination analysis in VLMs. K-Prism contains 1,002 multiple-choice questions across text and image tracks, organized into shared-evidence topic groups. Within each group, questions are designed to probe three diagnostic dimensions: factual, reasoning, and misleading. K-Prism uses a controlled dual-mode protocol comparing intact inputs with visually perturbed inputs to examine hallucination under reduced visual evidence. Experiments with closed- and open-source VLMs show dimension-specific degradation rather than a uniform accuracy drop. These findings demonstrate that hallucination in Korean VLMs cannot be captured by a single aggregate score and requires fine-grained diagnostic evaluation across evidence conditions and failure dimensions. Our dataset is publicly available at https://huggingface.co/spaces/KETI-NLP/K-Prism_Benchmark.
%U https://aclanthology.org/2026.aacl-short.41/
%P 632-644

Markdown (Informal)

[K-Prism: A Korean Diagnostic Benchmark for Multi-faceted Hallucination in Vision-Language Models](https://aclanthology.org/2026.aacl-short.41/) (Kim et al., AACL-IJCNLP 2026)

• K-Prism: A Korean Diagnostic Benchmark for Multi-faceted Hallucination in Vision-Language Models (Kim et al., AACL-IJCNLP 2026)

ACL

• Min Hyuk Kim, Dong Hyok Shin, Jimi Lee, and Jin Yea Jang. 2026. K-Prism: A Korean Diagnostic Benchmark for Multi-faceted Hallucination in Vision-Language Models. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers), pages 632–644, Hengqin, China. Association for Computational Linguistics.