<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ling-etal-2026-seeing">
    <titleInfo>
        <title>Seeing vs. Believing: Evaluating Open-Source MLLMs in Counter-Intuitive Scenes</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Chen</namePart>
        <namePart type="family">Ling</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tongwei</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hanqian</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Nai</namePart>
        <namePart type="family">Ding</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 (MLLMs) have demonstrated remarkable performance in mainstream visual understanding tasks, but their ability to process action scenes that contradict everyday common sense remains undertested. To address this gap, we introduce CAIT, a benchmark comprising 400 high-fidelity synthetic scenes focused on counter-intuitive visual actions, such as “a rabbit is chasing a tiger”, where visual evidence explicitly contradicts common-sense expectations. We evaluate human, leading proprietary models (e.g., Claude and Gemini), and 14 representative open-source MLLMs. Humans achieve near-perfect performance (around 0.95 accuracy) and proprietary models demonstrate robust understanding (achieving up to 0.88 accuracy), standard open-source instruction-tuned models perform at the chance level. Further analysis demonstrates that this failure is driven by a strong language prior: rather than trusting the visual input, they automatically override the anomalous visual signals with statistically common text descriptions. Although introducing Chain-of-Thought (CoT) reasoning mechanisms can improve accuracy, it significantly slows down the response and generates a new failure mode: models overthink the scenario and refuse to accept the actual visual content simply because it violates real-world physical laws. Finally, we demonstrate that targeted fine-tuning and structured prompting can effectively mitigate this reliance on language priors, enabling open-source models to accurately ground their reasoning in actual visual evidence.</abstract>
    <identifier type="citekey">ling-etal-2026-seeing</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.83/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>1695</start>
            <end>1712</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Seeing vs. Believing: Evaluating Open-Source MLLMs in Counter-Intuitive Scenes
%A Ling, Chen
%A Zhang, Tongwei
%A Li, Hanqian
%A Ding, Nai
%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 ling-etal-2026-seeing
%X Multimodal Large Language Models (MLLMs) have demonstrated remarkable performance in mainstream visual understanding tasks, but their ability to process action scenes that contradict everyday common sense remains undertested. To address this gap, we introduce CAIT, a benchmark comprising 400 high-fidelity synthetic scenes focused on counter-intuitive visual actions, such as “a rabbit is chasing a tiger”, where visual evidence explicitly contradicts common-sense expectations. We evaluate human, leading proprietary models (e.g., Claude and Gemini), and 14 representative open-source MLLMs. Humans achieve near-perfect performance (around 0.95 accuracy) and proprietary models demonstrate robust understanding (achieving up to 0.88 accuracy), standard open-source instruction-tuned models perform at the chance level. Further analysis demonstrates that this failure is driven by a strong language prior: rather than trusting the visual input, they automatically override the anomalous visual signals with statistically common text descriptions. Although introducing Chain-of-Thought (CoT) reasoning mechanisms can improve accuracy, it significantly slows down the response and generates a new failure mode: models overthink the scenario and refuse to accept the actual visual content simply because it violates real-world physical laws. Finally, we demonstrate that targeted fine-tuning and structured prompting can effectively mitigate this reliance on language priors, enabling open-source models to accurately ground their reasoning in actual visual evidence.
%U https://aclanthology.org/2026.findings-emnlp.83/
%P 1695-1712

Markdown (Informal)

[Seeing vs. Believing: Evaluating Open-Source MLLMs in Counter-Intuitive Scenes](https://aclanthology.org/2026.findings-emnlp.83/) (Ling et al., Findings 2026)

• Seeing vs. Believing: Evaluating Open-Source MLLMs in Counter-Intuitive Scenes (Ling et al., Findings 2026)

ACL

• Chen Ling, Tongwei Zhang, Hanqian Li, and Nai Ding. 2026. Seeing vs. Believing: Evaluating Open-Source MLLMs in Counter-Intuitive Scenes. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 1695–1712, Budapest, Hungary. Association for Computational Linguistics.