@inproceedings{zhao-etal-2026-improving,
title = "Improving Multimodal Contrastive Learning of Sentence Embeddings with Object-Phrase Alignment",
author = "Zhao, Kaiyan and
Miao, Zhongtao and
Tsuruoka, Yoshimasa",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-aacl.22/",
pages = "397--404",
ISBN = "979-8-89176-444-6",
abstract = "Multimodal sentence embedding models typically leverage image-caption pairs in addition to textual data during training. However, such pairs often contain noise, including redundant or irrelevant information on either the image or caption side. To mitigate this issue, we propose MCSEO, a method that enhances multimodal sentence embeddings by incorporating fine-grained object-phrase alignment alongside traditional image-caption alignment. Specifically, MCSEO utilizes existing detection and segmentation models to extract accurate object-phrase pairs, which are then used to optimize a proposed contrastive learning objective tailored to object-phrase correspondence. Experimental results on semantic textual similarity (STS) tasks across different backbone models demonstrate that MCSEO outperforms image-caption-only baselines on average, and that it is complementary to existing methods: combining MCSEO with KDMCSE yields further gains. A random-pairing control and a manual evaluation of the extracted pairs additionally indicate that the improvements stem from correct object-phrase alignment."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2026-improving">
<titleInfo>
<title>Improving Multimodal Contrastive Learning of Sentence Embeddings with Object-Phrase Alignment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kaiyan</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongtao</namePart>
<namePart type="family">Miao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoshimasa</namePart>
<namePart type="family">Tsuruoka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-444-6</identifier>
</relatedItem>
<abstract>Multimodal sentence embedding models typically leverage image-caption pairs in addition to textual data during training. However, such pairs often contain noise, including redundant or irrelevant information on either the image or caption side. To mitigate this issue, we propose MCSEO, a method that enhances multimodal sentence embeddings by incorporating fine-grained object-phrase alignment alongside traditional image-caption alignment. Specifically, MCSEO utilizes existing detection and segmentation models to extract accurate object-phrase pairs, which are then used to optimize a proposed contrastive learning objective tailored to object-phrase correspondence. Experimental results on semantic textual similarity (STS) tasks across different backbone models demonstrate that MCSEO outperforms image-caption-only baselines on average, and that it is complementary to existing methods: combining MCSEO with KDMCSE yields further gains. A random-pairing control and a manual evaluation of the extracted pairs additionally indicate that the improvements stem from correct object-phrase alignment.</abstract>
<identifier type="citekey">zhao-etal-2026-improving</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.22/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>397</start>
<end>404</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Multimodal Contrastive Learning of Sentence Embeddings with Object-Phrase Alignment
%A Zhao, Kaiyan
%A Miao, Zhongtao
%A Tsuruoka, Yoshimasa
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F zhao-etal-2026-improving
%X Multimodal sentence embedding models typically leverage image-caption pairs in addition to textual data during training. However, such pairs often contain noise, including redundant or irrelevant information on either the image or caption side. To mitigate this issue, we propose MCSEO, a method that enhances multimodal sentence embeddings by incorporating fine-grained object-phrase alignment alongside traditional image-caption alignment. Specifically, MCSEO utilizes existing detection and segmentation models to extract accurate object-phrase pairs, which are then used to optimize a proposed contrastive learning objective tailored to object-phrase correspondence. Experimental results on semantic textual similarity (STS) tasks across different backbone models demonstrate that MCSEO outperforms image-caption-only baselines on average, and that it is complementary to existing methods: combining MCSEO with KDMCSE yields further gains. A random-pairing control and a manual evaluation of the extracted pairs additionally indicate that the improvements stem from correct object-phrase alignment.
%U https://aclanthology.org/2026.findings-aacl.22/
%P 397-404
Markdown (Informal)
[Improving Multimodal Contrastive Learning of Sentence Embeddings with Object-Phrase Alignment](https://aclanthology.org/2026.findings-aacl.22/) (Zhao et al., Findings 2026)
ACL