@inproceedings{zhou-etal-2025-parameter,
title = "Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models",
author = "Zhou, Xingran and
Yang, Kun and
Miao, Changtao and
Hu, Bingyu and
Xu, Zhuoer and
Cui, Shiwen and
Meng, Changhua and
Hong, Dan",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.33/",
doi = "10.18653/v1/2025.naacl-long.33",
pages = "751--761",
ISBN = "979-8-89176-189-6",
abstract = "Large pre-trained Vision-Language Models (VLMs) have revolutionized both computer vision and natural language processing. Despite their success, adversarial examples can still mislead VLMs into producing incorrect results. This work focuses on boosting the adversarial robustness of VLMs by searching for text prompts at the word level, rather than optimizing continuous textual embeddings. We introduce Parameter-Free Prompt Tuning (PFPT) to learn defense words that enhance resilience against adversarial attacks when appended to existing prompts, thereby offering ease of use due to the simplicity of this approach. These defense words are naturally present in the inherent vocabulary of VLMs, providing a human-readable property. PFPT employs a coarse-to-fine strategy with carefully designed optimization objectives to guide the word search. Extensive experiments demonstrate our method{'}s superiority over hand-engineered prompts and other state-of-the-art methods. PFPT significantly boosts accuracy and robustness, outperforming hand-engineered prompts with average gains of +4.9{\%} and +5.8{\%}, respectively (epsilon=1/255)."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2025-parameter">
<titleInfo>
<title>Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xingran</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kun</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Changtao</namePart>
<namePart type="family">Miao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bingyu</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhuoer</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shiwen</namePart>
<namePart type="family">Cui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Changhua</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Hong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>Large pre-trained Vision-Language Models (VLMs) have revolutionized both computer vision and natural language processing. Despite their success, adversarial examples can still mislead VLMs into producing incorrect results. This work focuses on boosting the adversarial robustness of VLMs by searching for text prompts at the word level, rather than optimizing continuous textual embeddings. We introduce Parameter-Free Prompt Tuning (PFPT) to learn defense words that enhance resilience against adversarial attacks when appended to existing prompts, thereby offering ease of use due to the simplicity of this approach. These defense words are naturally present in the inherent vocabulary of VLMs, providing a human-readable property. PFPT employs a coarse-to-fine strategy with carefully designed optimization objectives to guide the word search. Extensive experiments demonstrate our method’s superiority over hand-engineered prompts and other state-of-the-art methods. PFPT significantly boosts accuracy and robustness, outperforming hand-engineered prompts with average gains of +4.9% and +5.8%, respectively (epsilon=1/255).</abstract>
<identifier type="citekey">zhou-etal-2025-parameter</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.33</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.33/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>751</start>
<end>761</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models
%A Zhou, Xingran
%A Yang, Kun
%A Miao, Changtao
%A Hu, Bingyu
%A Xu, Zhuoer
%A Cui, Shiwen
%A Meng, Changhua
%A Hong, Dan
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F zhou-etal-2025-parameter
%X Large pre-trained Vision-Language Models (VLMs) have revolutionized both computer vision and natural language processing. Despite their success, adversarial examples can still mislead VLMs into producing incorrect results. This work focuses on boosting the adversarial robustness of VLMs by searching for text prompts at the word level, rather than optimizing continuous textual embeddings. We introduce Parameter-Free Prompt Tuning (PFPT) to learn defense words that enhance resilience against adversarial attacks when appended to existing prompts, thereby offering ease of use due to the simplicity of this approach. These defense words are naturally present in the inherent vocabulary of VLMs, providing a human-readable property. PFPT employs a coarse-to-fine strategy with carefully designed optimization objectives to guide the word search. Extensive experiments demonstrate our method’s superiority over hand-engineered prompts and other state-of-the-art methods. PFPT significantly boosts accuracy and robustness, outperforming hand-engineered prompts with average gains of +4.9% and +5.8%, respectively (epsilon=1/255).
%R 10.18653/v1/2025.naacl-long.33
%U https://aclanthology.org/2025.naacl-long.33/
%U https://doi.org/10.18653/v1/2025.naacl-long.33
%P 751-761
Markdown (Informal)
[Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models](https://aclanthology.org/2025.naacl-long.33/) (Zhou et al., NAACL 2025)
ACL
- Xingran Zhou, Kun Yang, Changtao Miao, Bingyu Hu, Zhuoer Xu, Shiwen Cui, Changhua Meng, and Dan Hong. 2025. Parameter-free and Accessible Prompt Learning to Enhance Adversarial Robustness for Pre-trained Vision-Language Models. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 751–761, Albuquerque, New Mexico. Association for Computational Linguistics.