BadBone: Backdoor Attacks Against Backbone Models in Visual Prompt Learning
- 1 June 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 5530-5543
Abstract
Prompt learning is a new machine learning paradigm that has attracted ample attention due to its simplicity and proven efficacy. Despite its growing adoption, the security vulnerabilities associated with this paradigm remain underexplored. In this work, we take the first step to propose BadBone, a stealthy and adaptive backdoor attack against prompt learning using bi-level optimization. Instead of backdooring the prompt learning process, we aim to compromise a backbone model such that only target downstream tasks employing prompt learning inherit the backdoor vulnerability. Extensive experiments on three different models and three datasets from various domains show that our targeted/untargeted backdoored models achieve high attack performance while maintaining utility on both pretraining and downstream tasks. Moreover, we evaluate our approach against six state-of-the-art model-level defenses, including Neural Cleanse, ABS, MNTD, NAD, CLP, and D-BR. The results demonstrate that these defenses are largely ineffective against our backdoored models and thus leave the effective defense as an important direction for future work. Our code is available at https://github.com/TrustAIRLab/BadBone.Keywords
All Related Versions
Funding Information
- National Key Research and Development Program of China (2025YFB3110200)
- Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research (26K21215)
- Basic and Applied Basic Research Foundation of Guangdong Province (2026A1515030046)
- State Key Laboratory of Internet Architecture (HLW2025ZD14)
This publication has 42 references indexed in Scilit:
- Data-Free Backdoor Removal Based on Channel Lipschitzness2022
- Visual Prompt Tuning2022
- VulnerGAN: a backdoor attack through vulnerability amplification against machine learning-based network intrusion detection systemsScience China Information Sciences, 2022
- PTR: Prompt Tuning with Rules for Text ClassificationAI Open, 2022
- Defending Deep Neural Networks Against Backdoor Attack by Using De-Trigger AutoencoderIEEE Access, 2021
- Big Transfer (BiT): General Visual Representation Learning2020
- Clean-Label Backdoor Attacks on Video Recognition Models2020
- Age Progression/Regression by Conditional Adversarial Autoencoder2017
- Deep Residual Learning for Image Recognition2016
- ImageNet: A large-scale hierarchical image database2009