<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-skill-augmented">
    <titleInfo>
        <title>A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yue</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Liqiang</namePart>
        <namePart type="family">Jing</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jia</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yapeng</namePart>
        <namePart type="family">Tian</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xinya</namePart>
        <namePart type="family">Du</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yunhui</namePart>
        <namePart type="family">Guo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Vibhav</namePart>
        <namePart type="given">Giridhar</namePart>
        <namePart type="family">Gogate</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 Large Language Models have achieved strong performance in single-video understanding, yet their ability to reason across multiple videos remains limited. Existing approaches typically concatenate multiple videos into a single input and perform direct inference, which introduces training–inference mismatch, information loss from frame compression, and a lack of explicit cross-video coordination. Meanwhile, current multi-video benchmarks primarily emphasize event-level comparison, leaving identity-level matching, fine-grained discrimination, and structured multi-step reasoning underexplored. To address these gaps, we introduce MVX-Bench, a Multi-Video Cross-Dimension Benchmark that reformulates 11 classical computer vision tasks into a unified multi-video question-answering framework, comprising 1,442 questions over 4,255 videos from diverse real-world datasets. We further propose SAMA, a Skill-Augmented Agentic Framework for Multi-Video Understanding, which integrates visual tools, task-specific skills, and a conflict-aware verification mechanism to support iterative and structured reasoning. Experiments show that SAMA improves substantially over its visual perception backbone and achieves the best average accuracy on MVX-Bench. Ablations confirm the benefits of skill-guided orchestration and conflict-aware verification.</abstract>
    <identifier type="citekey">zhang-etal-2026-skill-augmented</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.429/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>9032</start>
            <end>9057</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding
%A Zhang, Yue
%A Jing, Liqiang
%A Li, Jia
%A Tian, Yapeng
%A Du, Xinya
%A Guo, Yunhui
%A Gogate, Vibhav Giridhar
%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-skill-augmented
%X Multimodal Large Language Models have achieved strong performance in single-video understanding, yet their ability to reason across multiple videos remains limited. Existing approaches typically concatenate multiple videos into a single input and perform direct inference, which introduces training–inference mismatch, information loss from frame compression, and a lack of explicit cross-video coordination. Meanwhile, current multi-video benchmarks primarily emphasize event-level comparison, leaving identity-level matching, fine-grained discrimination, and structured multi-step reasoning underexplored. To address these gaps, we introduce MVX-Bench, a Multi-Video Cross-Dimension Benchmark that reformulates 11 classical computer vision tasks into a unified multi-video question-answering framework, comprising 1,442 questions over 4,255 videos from diverse real-world datasets. We further propose SAMA, a Skill-Augmented Agentic Framework for Multi-Video Understanding, which integrates visual tools, task-specific skills, and a conflict-aware verification mechanism to support iterative and structured reasoning. Experiments show that SAMA improves substantially over its visual perception backbone and achieves the best average accuracy on MVX-Bench. Ablations confirm the benefits of skill-guided orchestration and conflict-aware verification.
%U https://aclanthology.org/2026.findings-emnlp.429/
%P 9032-9057

Markdown (Informal)

[A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding](https://aclanthology.org/2026.findings-emnlp.429/) (Zhang et al., Findings 2026)

• A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding (Zhang et al., Findings 2026)

ACL

• Yue Zhang, Liqiang Jing, Jia Li, Yapeng Tian, Xinya Du, Yunhui Guo, and Vibhav Giridhar Gogate. 2026. A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 9032–9057, Budapest, Hungary. Association for Computational Linguistics.