<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-divide">
    <titleInfo>
        <title>Divide-and-Adapt: Fine-Tuning VLMs via Model Division and Routing</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hailing</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Neil</namePart>
        <namePart type="family">Rodrigues</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xianglong</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wei</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lin</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yun</namePart>
        <namePart type="family">Fu</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>Fine-tuning Vision-Language Models (VLMs) is a common practice to adapt pre-trained models for downstream tasks. Yet existing fine-tuning methods consistently keep the pre-trained architecture intact, leaving open the question of whether maintaining such structural integrity is truly necessary for effective adaptation. To investigate this, we propose Divide-and-Adapt, a new fine-tuning paradigm that departs from preserving the full pre-trained structure. It first divides the pre-trained model into multiple downstream-oriented subnetworks, and then adapts them collaboratively through a routing-based coordination mechanism. This design facilitates the selection of structures that are most beneficial for downstream tasks and enables their collaborative integration. Remarkably, even without any task-specific training, our Divide-and-Adapt framework first divides the pre-trained model into downstream-oriented subnetworks and adaptively activates them for downstream inference, already outperforming the pre-trained model on downstream tasks. With further fine-tuning, Divide-and-Adapt surpasses both full fine-tuning and parameter-efficient baselines across multiple benchmarks, demonstrating its effectiveness. Our code will be made publicly available at https://github.com/haiiling/DivideAndAdapt.</abstract>
    <identifier type="citekey">wang-etal-2026-divide</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.615/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>14137</start>
            <end>14151</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Divide-and-Adapt: Fine-Tuning VLMs via Model Division and Routing
%A Wang, Hailing
%A Rodrigues, Neil
%A Wang, Xianglong
%A Li, Wei
%A Chen, Lin
%A Fu, Yun
%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-divide
%X Fine-tuning Vision-Language Models (VLMs) is a common practice to adapt pre-trained models for downstream tasks. Yet existing fine-tuning methods consistently keep the pre-trained architecture intact, leaving open the question of whether maintaining such structural integrity is truly necessary for effective adaptation. To investigate this, we propose Divide-and-Adapt, a new fine-tuning paradigm that departs from preserving the full pre-trained structure. It first divides the pre-trained model into multiple downstream-oriented subnetworks, and then adapts them collaboratively through a routing-based coordination mechanism. This design facilitates the selection of structures that are most beneficial for downstream tasks and enables their collaborative integration. Remarkably, even without any task-specific training, our Divide-and-Adapt framework first divides the pre-trained model into downstream-oriented subnetworks and adaptively activates them for downstream inference, already outperforming the pre-trained model on downstream tasks. With further fine-tuning, Divide-and-Adapt surpasses both full fine-tuning and parameter-efficient baselines across multiple benchmarks, demonstrating its effectiveness. Our code will be made publicly available at https://github.com/haiiling/DivideAndAdapt.
%U https://aclanthology.org/2026.emnlp-main.615/
%P 14137-14151

Markdown (Informal)

[Divide-and-Adapt: Fine-Tuning VLMs via Model Division and Routing](https://aclanthology.org/2026.emnlp-main.615/) (Wang et al., EMNLP 2026)

• Divide-and-Adapt: Fine-Tuning VLMs via Model Division and Routing (Wang et al., EMNLP 2026)

ACL

• Hailing Wang, Neil Rodrigues, Xianglong Wang, Wei Li, Lin Chen, and Yun Fu. 2026. Divide-and-Adapt: Fine-Tuning VLMs via Model Division and Routing. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 14137–14151, Budapest, Hungary. Association for Computational Linguistics.