<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-vision">
    <titleInfo>
        <title>Do Vision Language Models Understand Human Engagement in Games?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ziyi</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qizan</namePart>
        <namePart type="family">Guo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Rishitosh</namePart>
        <namePart type="given">Kumar</namePart>
        <namePart type="family">Singh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiyang</namePart>
        <namePart type="family">Hu</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Inferring human engagement from gameplay video is important for game design and player-experience research, yet it remains unclear whether vision–language models (VLMs) can infer such latent psychological states from visual cues alone. Using the GameVibe Few-Shot dataset across nine first-person shooter games, we evaluate three VLMs under six prompting strategies, including zero-shot prediction, theory-guided prompts grounded in Flow, GameFlow, Self-Determination Theory, and MDA, and retrieval-augmented prompting. We consider both pointwise engagement prediction and pairwise prediction of engagement change between consecutive windows. Results show that zero-shot VLM predictions are generally weak and often fail to outperform simple per-game majority-class baselines. Memory- or retrieval-augmented prompting improves pointwise prediction in some settings, whereas pairwise prediction remains consistently difficult across strategies. Theory-guided prompting alone does not reliably help and can instead reinforce surface-level shortcuts. These findings suggest a perception–understanding gap in current VLMs: although they can recognize visible gameplay cues, they still struggle to robustly infer human engagement across games.</abstract>
    <identifier type="citekey">wang-etal-2026-vision</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1636/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>37184</start>
            <end>37201</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Do Vision Language Models Understand Human Engagement in Games?
%A Wang, Ziyi
%A Guo, Qizan
%A Singh, Rishitosh Kumar
%A Hu, Xiyang
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F wang-etal-2026-vision
%X Inferring human engagement from gameplay video is important for game design and player-experience research, yet it remains unclear whether vision–language models (VLMs) can infer such latent psychological states from visual cues alone. Using the GameVibe Few-Shot dataset across nine first-person shooter games, we evaluate three VLMs under six prompting strategies, including zero-shot prediction, theory-guided prompts grounded in Flow, GameFlow, Self-Determination Theory, and MDA, and retrieval-augmented prompting. We consider both pointwise engagement prediction and pairwise prediction of engagement change between consecutive windows. Results show that zero-shot VLM predictions are generally weak and often fail to outperform simple per-game majority-class baselines. Memory- or retrieval-augmented prompting improves pointwise prediction in some settings, whereas pairwise prediction remains consistently difficult across strategies. Theory-guided prompting alone does not reliably help and can instead reinforce surface-level shortcuts. These findings suggest a perception–understanding gap in current VLMs: although they can recognize visible gameplay cues, they still struggle to robustly infer human engagement across games.
%U https://aclanthology.org/2026.emnlp-main.1636/
%P 37184-37201

Markdown (Informal)

[Do Vision Language Models Understand Human Engagement in Games?](https://aclanthology.org/2026.emnlp-main.1636/) (Wang et al., EMNLP 2026)

• Do Vision Language Models Understand Human Engagement in Games? (Wang et al., EMNLP 2026)

ACL

• Ziyi Wang, Qizan Guo, Rishitosh Kumar Singh, and Xiyang Hu. 2026. Do Vision Language Models Understand Human Engagement in Games?. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 37184–37201, Budapest, Hungary. Association for Computational Linguistics.