<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2025-transferring">
    <titleInfo>
        <title>Transferring Textual Preferences to Vision-Language Understanding through Model Merging</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Chen-An</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tzu-Han</namePart>
        <namePart type="family">Lin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yun-Nung</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hung-yi</namePart>
        <namePart type="family">Lee</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Wanxiang</namePart>
            <namePart type="family">Che</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joyce</namePart>
            <namePart type="family">Nabende</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ekaterina</namePart>
            <namePart type="family">Shutova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohammad</namePart>
            <namePart type="given">Taher</namePart>
            <namePart type="family">Pilehvar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Vienna, Austria</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-252-7</identifier>
    </relatedItem>
    <abstract>Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs’ scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs.</abstract>
    <identifier type="citekey">li-etal-2025-transferring</identifier>
    <identifier type="doi">10.18653/v1/2025.acl-short.72</identifier>
    <location>
        <url>https://aclanthology.org/2025.acl-short.72/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>923</start>
            <end>943</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Transferring Textual Preferences to Vision-Language Understanding through Model Merging
%A Li, Chen-An
%A Lin, Tzu-Han
%A Chen, Yun-Nung
%A Lee, Hung-yi
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-252-7
%F li-etal-2025-transferring
%X Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs’ scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs.
%R 10.18653/v1/2025.acl-short.72
%U https://aclanthology.org/2025.acl-short.72/
%U https://doi.org/10.18653/v1/2025.acl-short.72
%P 923-943

Markdown (Informal)

[Transferring Textual Preferences to Vision-Language Understanding through Model Merging](https://aclanthology.org/2025.acl-short.72/) (Li et al., ACL 2025)

• Transferring Textual Preferences to Vision-Language Understanding through Model Merging (Li et al., ACL 2025)

ACL

• Chen-An Li, Tzu-Han Lin, Yun-Nung Chen, and Hung-yi Lee. 2025. Transferring Textual Preferences to Vision-Language Understanding through Model Merging. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 923–943, Vienna, Austria. Association for Computational Linguistics.