Adversarial Backdoor

Adversarial backdoor attacks exploit vulnerabilities in machine learning models by embedding hidden triggers that cause misclassification without significantly impacting performance on clean data. Current research focuses on developing both more sophisticated attacks, including those leveraging generative models and targeting diverse model architectures like LLMs and those used in autonomous driving, and more robust defenses, such as those employing data augmentation techniques, model editing, and contributor-aware approaches. The significance of this research lies in its implications for the security and reliability of AI systems across various applications, from facial recognition and autonomous vehicles to federated learning and large language models.

Papers

October 15, 2024

AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment
Pankayaraj Pathmanathan, Udari Madhushani Sehwag, Michael-Andrei Panaitescu-Liess, Furong Huang
Backdoor Attack Backdoor Trigger Synchronous Generator LLM Alignment Prompt Attack Robust Defense Adversarial Backdoor

September 24, 2024

Adversarial Backdoor Defense in CLIP
Junhao Kuang, Siyuan Liang, Jiawei Liang, Kuanrong Liu, Xiaochun Cao
Adversarial Example Single CLIP Backdoor Defense Backdoor Sample Adversarial Backdoor Multimodal Backdoor

August 27, 2024

Adversarial Manhole: Challenging Monocular Depth Estimation and Semantic Segmentation Models with Patch Attack
Naufal Suryanto, Andro Aprila Adiputra, Ahmada Yusril Kadiptya, Yongsu Kim, Howon Kim
Semantic Segmentation Monocular Depth Estimation Depth Map Semantic Segmentation Model Adversarial Patch Adversarial Scenario Patch Attack Adversarial Backdoor

August 22, 2024

MakeupAttack: Feature Space Black-box Backdoor Attack on Face Recognition via Makeup Transfer
Ming Sun, Lihua Jing, Zixuan Zhu, Rui Wang
Face Recognition Makeup Product Image Facial Datasets Makeup Transfer Adversarial Backdoor Space Attack

August 20, 2024

MEGen: Generative Backdoor in Large Language Models via Model Editing
Jiyang Qiu, Xinbei Ma, Zhuosheng Zhang, Hai Zhao
Large Language Model Language Model Backdoor Attack Adversarial Backdoor

May 9, 2024

Towards Robust Physical-world Backdoor Attacks on Lane Detection
Xinwei Zhang, Aishan Liu, Tianyuan Zhang, Siyuan Liang, Xianglong Liu
Autonomous Driving Dynamic Scene Lane Detection Backdoor Injection Adversarial Backdoor Physical Backdoor Attack

July 28, 2023

Beating Backdoor Attack at Its Own Game
Min Liu, Alberto Sangiovanni-Vincentelli, Xiangyu Yue
Backdoor Attack Adversarial Backdoor Mitigating Backdoor

July 1, 2023

Fedward: Flexible Federated Backdoor Defense Framework with Non-IID Data
Zekai Chen, Fuyi Wang, Zhiwei Zheng, Ximeng Liu, Yujie Lin
Raw Data Backdoor Attack Adversarial Backdoor Attacker Model

June 27, 2023

Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Driving
Mozhgan Pourkeshavarz, Mohammad Sabokrou, Amir Rasouli
Adversarial Attack Autonomous Driving Trajectory Prediction Trajectory Prediction Model Adversarial Backdoor

April 27, 2023

Adversary Aware Continual Learning
Muhammad Umer, Robi Polikar
Incremental Learning State of the Art Defense Adversarial Backdoor Continual Adversarial Defense

April 22, 2023

Universal Adversarial Backdoor Attacks to Fool Vertical Federated Learning in Cloud-Edge Collaboration
Peng Chen, Xin Du, Zhihui Lu, Hongfeng Chai
Backdoor Attack Vertical Federated Learning Clean Label Backdoor Attack Adversarial Backdoor Backdoor Behavior Edge Cloud Collaboration

October 17, 2022

Marksman Backdoor: Backdoor Attacks with Arbitrary Target Class
Khoa D. Doan, Yingjie Lao, Ping Li
Backdoor Attack Backdoor Defense Adversarial Backdoor Target Class

July 1, 2022

BadHash: Invisible Backdoor Attacks against Deep Hashing with Clean Label
Shengshan Hu, Ziqi Zhou, Yechao Zhang, Leo Yu Zhang, Yifeng Zheng, Yuanyuan HE, Hai Jin
Adversarial Attack Backdoor Attack Deep Hashing Adversarial Backdoor Invisible Backdoor Attack Clean Label Image Backdoor Attack

June 12, 2022

Neurotoxin: Durable Backdoors in Federated Learning
Zhengming Zhang, Ashwinee Panda, Linyue Song, Yaoqing Yang, Michael W. Mahoney, Joseph E. Gonzalez, Kannan Ramchandran, Prateek Mittal
Backdoor Attack Adversarial Backdoor Resilient Backdoor

May 28, 2022

Contributor-Aware Defenses Against Adversarial Backdoor Attacks
Glenn Dawson, Muhammad Umer, Robi Polikar
Adversarial Attack Multi Step Adversarial Attack Adversarial Backdoor Adversarial Trigger Certified Defense