@inproceedings{hu-etal-2026-unsupervised,
title = "Unsupervised Domain Specialization for Multimodal Embeddings",
author = "Hu, Ruifeng and
Lian, Jianxun and
Wu, Mingqi and
Zhou, Mingyang and
Liao, Hao and
Chao, Naipeng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1227/",
doi = "10.18653/2026.findings-emnlp.1227",
pages = "25976--25994",
ISBN = "979-8-89176-462-0",
abstract = "Recent advances in multimodal foundation models have enabled powerful generic embeddings. However, real-world applications{---}such as e-commerce or healthcare{---}rely on domain-specific patterns that these generic models often fail to capture precisely. Adapting them without readily available labeled data remains a critical challenge. In this paper, we propose MM-UDA, a robust unsupervised domain adaptation framework designed to systematically convert a generic multimodal embedding model into a domain expert using only unlabeled target data. Our framework employs a two-stage strategy. First, we introduce a within-modality pseudo-labeling task refined by Gaussian Mixture Model filtering, which rapidly adapts the model to domain-specific features while suppressing label noise. Then, we align modalities through a score-guided learning scheme to refine cross-modal pairing. Extensive evaluations on three real-world datasets demonstrate that MM-UDA both significantly enhances various backbone models and consistently outperforms competitive baselines, confirming its effectiveness for cross-modal retrieval in specialized domains. The source code is publicly available at https://github.com/hrf516/MM-UDA."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hu-etal-2026-unsupervised">
<titleInfo>
<title>Unsupervised Domain Specialization for Multimodal Embeddings</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianxun</namePart>
<namePart type="family">Lian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingqi</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyang</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Liao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naipeng</namePart>
<namePart type="family">Chao</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>Recent advances in multimodal foundation models have enabled powerful generic embeddings. However, real-world applications—such as e-commerce or healthcare—rely on domain-specific patterns that these generic models often fail to capture precisely. Adapting them without readily available labeled data remains a critical challenge. In this paper, we propose MM-UDA, a robust unsupervised domain adaptation framework designed to systematically convert a generic multimodal embedding model into a domain expert using only unlabeled target data. Our framework employs a two-stage strategy. First, we introduce a within-modality pseudo-labeling task refined by Gaussian Mixture Model filtering, which rapidly adapts the model to domain-specific features while suppressing label noise. Then, we align modalities through a score-guided learning scheme to refine cross-modal pairing. Extensive evaluations on three real-world datasets demonstrate that MM-UDA both significantly enhances various backbone models and consistently outperforms competitive baselines, confirming its effectiveness for cross-modal retrieval in specialized domains. The source code is publicly available at https://github.com/hrf516/MM-UDA.</abstract>
<identifier type="citekey">hu-etal-2026-unsupervised</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.1227</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1227/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>25976</start>
<end>25994</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Unsupervised Domain Specialization for Multimodal Embeddings
%A Hu, Ruifeng
%A Lian, Jianxun
%A Wu, Mingqi
%A Zhou, Mingyang
%A Liao, Hao
%A Chao, Naipeng
%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 hu-etal-2026-unsupervised
%X Recent advances in multimodal foundation models have enabled powerful generic embeddings. However, real-world applications—such as e-commerce or healthcare—rely on domain-specific patterns that these generic models often fail to capture precisely. Adapting them without readily available labeled data remains a critical challenge. In this paper, we propose MM-UDA, a robust unsupervised domain adaptation framework designed to systematically convert a generic multimodal embedding model into a domain expert using only unlabeled target data. Our framework employs a two-stage strategy. First, we introduce a within-modality pseudo-labeling task refined by Gaussian Mixture Model filtering, which rapidly adapts the model to domain-specific features while suppressing label noise. Then, we align modalities through a score-guided learning scheme to refine cross-modal pairing. Extensive evaluations on three real-world datasets demonstrate that MM-UDA both significantly enhances various backbone models and consistently outperforms competitive baselines, confirming its effectiveness for cross-modal retrieval in specialized domains. The source code is publicly available at https://github.com/hrf516/MM-UDA.
%R 10.18653/2026.findings-emnlp.1227
%U https://aclanthology.org/2026.findings-emnlp.1227/
%U https://doi.org/10.18653/2026.findings-emnlp.1227
%P 25976-25994
Markdown (Informal)
[Unsupervised Domain Specialization for Multimodal Embeddings](https://aclanthology.org/2026.findings-emnlp.1227/) (Hu et al., Findings 2026)
ACL
- Ruifeng Hu, Jianxun Lian, Mingqi Wu, Mingyang Zhou, Hao Liao, and Naipeng Chao. 2026. Unsupervised Domain Specialization for Multimodal Embeddings. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 25976–25994, Budapest, Hungary. Association for Computational Linguistics.