@inproceedings{tang-etal-2023-xtremeclip,
title = "{X}treme{CLIP}: Extremely Parameter-efficient Tuning for Low-resource Vision Language Understanding",
author = "Tang, Moming and
Wang, Chengyu and
Wang, Jianing and
Tan, Chuanqi and
Huang, Songfang and
Chen, Cen and
Qian, Weining",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.397",
doi = "10.18653/v1/2023.findings-acl.397",
pages = "6368--6376",
abstract = "Recently, Contrastive Visual-Language Pre-training (CLIP) has demonstrated remarkable capability in various Visual Language Understanding (VLU) tasks. Yet, most CLIP-based methods require tasks-specific designs and sufficient training data. In this paper, we introduce a simple yet efficient paradigm for low-resource VLU named XtremeCLIP, which involves very few trainable parameters to improve the generalization ability of the trained models. In our XtremeCLIP framework, we reformulate a series of VLU tasks as a unified open-book affinity-matching problem. Furthermore, to handle the insufficient supervised signals in small datasets, we adopt contrastive learning to utilize the implicit sorting information of ground-truth labels to provide more supervised cues. Extensive experiments over multiple datasets on visual entailment, visual question answering, and image classification show that XtremeCLIP consistently outperforms existing baselines in low-resource settings.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tang-etal-2023-xtremeclip">
<titleInfo>
<title>XtremeCLIP: Extremely Parameter-efficient Tuning for Low-resource Vision Language Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Moming</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengyu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chuanqi</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Songfang</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weining</namePart>
<namePart type="family">Qian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recently, Contrastive Visual-Language Pre-training (CLIP) has demonstrated remarkable capability in various Visual Language Understanding (VLU) tasks. Yet, most CLIP-based methods require tasks-specific designs and sufficient training data. In this paper, we introduce a simple yet efficient paradigm for low-resource VLU named XtremeCLIP, which involves very few trainable parameters to improve the generalization ability of the trained models. In our XtremeCLIP framework, we reformulate a series of VLU tasks as a unified open-book affinity-matching problem. Furthermore, to handle the insufficient supervised signals in small datasets, we adopt contrastive learning to utilize the implicit sorting information of ground-truth labels to provide more supervised cues. Extensive experiments over multiple datasets on visual entailment, visual question answering, and image classification show that XtremeCLIP consistently outperforms existing baselines in low-resource settings.</abstract>
<identifier type="citekey">tang-etal-2023-xtremeclip</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.397</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.397</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>6368</start>
<end>6376</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T XtremeCLIP: Extremely Parameter-efficient Tuning for Low-resource Vision Language Understanding
%A Tang, Moming
%A Wang, Chengyu
%A Wang, Jianing
%A Tan, Chuanqi
%A Huang, Songfang
%A Chen, Cen
%A Qian, Weining
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F tang-etal-2023-xtremeclip
%X Recently, Contrastive Visual-Language Pre-training (CLIP) has demonstrated remarkable capability in various Visual Language Understanding (VLU) tasks. Yet, most CLIP-based methods require tasks-specific designs and sufficient training data. In this paper, we introduce a simple yet efficient paradigm for low-resource VLU named XtremeCLIP, which involves very few trainable parameters to improve the generalization ability of the trained models. In our XtremeCLIP framework, we reformulate a series of VLU tasks as a unified open-book affinity-matching problem. Furthermore, to handle the insufficient supervised signals in small datasets, we adopt contrastive learning to utilize the implicit sorting information of ground-truth labels to provide more supervised cues. Extensive experiments over multiple datasets on visual entailment, visual question answering, and image classification show that XtremeCLIP consistently outperforms existing baselines in low-resource settings.
%R 10.18653/v1/2023.findings-acl.397
%U https://aclanthology.org/2023.findings-acl.397
%U https://doi.org/10.18653/v1/2023.findings-acl.397
%P 6368-6376
Markdown (Informal)
[XtremeCLIP: Extremely Parameter-efficient Tuning for Low-resource Vision Language Understanding](https://aclanthology.org/2023.findings-acl.397) (Tang et al., Findings 2023)
ACL