Effective Backdoor

Effective backdoor attacks exploit vulnerabilities in machine learning models, surreptitiously manipulating their behavior by injecting malicious data or modifying model architecture. Current research focuses on understanding and mitigating these attacks across various model types, including large vision-language models, embodied AI systems, and federated learning architectures, exploring diverse attack vectors such as data poisoning, trigger injection (including subtle frequency-domain manipulations and even simple rotations), and circuit-level modifications for quantum neural networks. The widespread impact of these attacks on the security and reliability of AI systems, particularly in safety-critical applications like autonomous driving and robotics, necessitates ongoing research into robust defenses and detection methods.

Papers

June 27, 2024

Revisiting Backdoor Attacks against Large Vision-Language Models
Siyuan Liang, Jiawei Liang, Tianyu Pang, Chao Du, Aishan Liu, Ee-Chien Chang, Xiaochun Cao
Backdoor Attack Large Vision Language Model Transferable Attack Effective Backdoor

May 27, 2024

Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems
Ruochen Jiao, Shaoyuan Xie, Justin Yue, Takami Sato, Lixu Wang, Yixuan Wang, Qi Alfred Chen, Qi Zhu
Backdoor Attack Large Language Knowledge Injection LLM Assisted Decision Making Effective Backdoor

April 30, 2024

Let's Focus: Focused Backdoor Attack against Federated Transfer Learning
Marco Arazzi, Stefanos Koffas, Antonino Nocera, Stjepan Picek
Federated Learning Backdoor Attack Federated Transfer Learning Effective Backdoor

April 17, 2024

Detector Collapse: Physical-World Backdooring Object Detection to Catastrophic Overload or Blindness in Autonomous Driving
Hangtao Zhang, Shengshan Hu, Yichen Wang, Leo Yu Zhang, Ziqi Zhou, Xianlong Wang, Yanjun Zhang, Chao Chen
Backdoor Attack Backdoor Defense Shi Detector Blind User Effective Backdoor

March 11, 2024

AS-FIBA: Adaptive Selective Frequency-Injection for Backdoor Attack on Deep Face Restoration
Zhenbo Song, Wenhao Gao, Kaihao Zhang, Wenhan Luo, Zhaoxin Fan, Jianfeng Lu
Backdoor Attack Face Restoration Effective Backdoor

July 19, 2023

Rethinking Backdoor Attacks
Alaa Khaddaj, Guillaume Leclerc, Aleksandar Makelov, Kristian Georgiev, Hadi Salman, Andrew Ilyas, Aleksander Madry
Backdoor Attack Adversary Agent Effective Backdoor

February 16, 2023

QTrojan: A Circuit Backdoor Against Quantum Neural Networks
Cheng Chu, Lei Jiang, Martin Swany, Fan Chen
Backdoor Attack Quantum Neural Network Variational Quantum Circuit Quantum Gate Effective Backdoor

July 22, 2022

Just Rotate it: Deploying Backdoor Attacks via Rotation Transformation
Tong Wu, Tianhao Wang, Vikash Sehwag, Saeed Mahloujifar, Prateek Mittal
Backdoor Attack Backdoor Defense Attack Success Rate Backdoor Poisoning Attack Effective Backdoor