<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="corti-yang-2026-diavlo">
    <titleInfo>
        <title>DiaVLo: Diagnosing Behaviours of Vision-Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Lorenzo</namePart>
        <namePart type="family">Corti</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jie</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>Vision-language models (VLMs) rely on storing and transferring appropriate information across their sub-components. Verifying that the VLMs exhibit desired behaviours, while avoiding harmful ones, is central to their reliable deployment. Yet, methods that identify VLM behaviours remain scarce. We present DiaVLo, a diagnostic framework that leverages human curation and VLMs’ generation capabilities to construct specifications of desired and observed VLM behaviours, surfacing potential misalignments. Beyond this, DiaVLo also provides causal estimates to identify the most influential concepts steering VLM behaviours. We evaluate DiaVLo on several open-source VLMs under both classification and generation conditions. Our experiments show that DiaVLo produces behaviour labels that correlate with model performance and provide context for measured performance. DiaVLo surfaced behaviours that are clearly aligned and misaligned, alongside patterns in how VLMs perceive, organise, and prioritise concepts.</abstract>
    <identifier type="citekey">corti-yang-2026-diavlo</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.1841</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1841/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>39299</start>
            <end>39332</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T DiaVLo: Diagnosing Behaviours of Vision-Language Models
%A Corti, Lorenzo
%A Yang, Jie
%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 corti-yang-2026-diavlo
%X Vision-language models (VLMs) rely on storing and transferring appropriate information across their sub-components. Verifying that the VLMs exhibit desired behaviours, while avoiding harmful ones, is central to their reliable deployment. Yet, methods that identify VLM behaviours remain scarce. We present DiaVLo, a diagnostic framework that leverages human curation and VLMs’ generation capabilities to construct specifications of desired and observed VLM behaviours, surfacing potential misalignments. Beyond this, DiaVLo also provides causal estimates to identify the most influential concepts steering VLM behaviours. We evaluate DiaVLo on several open-source VLMs under both classification and generation conditions. Our experiments show that DiaVLo produces behaviour labels that correlate with model performance and provide context for measured performance. DiaVLo surfaced behaviours that are clearly aligned and misaligned, alongside patterns in how VLMs perceive, organise, and prioritise concepts.
%R 10.18653/2026.findings-emnlp.1841
%U https://aclanthology.org/2026.findings-emnlp.1841/
%U https://doi.org/10.18653/2026.findings-emnlp.1841
%P 39299-39332

Markdown (Informal)

[DiaVLo: Diagnosing Behaviours of Vision-Language Models](https://aclanthology.org/2026.findings-emnlp.1841/) (Corti & Yang, Findings 2026)

• DiaVLo: Diagnosing Behaviours of Vision-Language Models (Corti & Yang, Findings 2026)

ACL

• Lorenzo Corti and Jie Yang. 2026. DiaVLo: Diagnosing Behaviours of Vision-Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 39299–39332, Budapest, Hungary. Association for Computational Linguistics.