Imperceptible Trigger

Imperceptible triggers are covert manipulations embedded in machine learning models, particularly deep neural networks, to induce unintended behavior without noticeably altering the model's performance on clean data. Current research focuses on developing increasingly stealthy triggers for various model types, including concept bottleneck models and self-supervised learning architectures, and on improving attack effectiveness while minimizing trigger detectability. This area is crucial for enhancing the security and trustworthiness of AI systems across diverse applications, from image classification and speech recognition to network security and reinforcement learning, by identifying and mitigating vulnerabilities to malicious manipulation.

Papers

October 7, 2024

CAT: Concept-level backdoor ATtacks for Concept Bottleneck Models
Songning Lai, Jiayu Yang, Yu Huang, Lijie Hu, Tianlang Xue, Zhangyi Hu, Jiaxu Li, Haicheng Liao, Yutao Yue
Backdoor Attack Concept Bottleneck Model 2 Cat Imperceptible Trigger

August 12, 2024

Animate, or Inanimate, That is the Question for Large Language Models
Leonardo Ranaldi, Giulia Pucci, Fabio Massimo Zanzotto
Natural Language Yes No Question Human Like Mobile Object Imperceptible Trigger Parallel Personification Data

May 23, 2024

Towards Imperceptible Backdoor Attack in Self-supervised Learning
Hanrong Zhang, Zhenting Wang, Tingxu Han, Mingyu Jin, Chenlu Zhan, Mengnan Du, Hongwei Wang, Shiqing Ma
Self Supervised Learning Backdoor Attack Backdoor Defense Effective Backdoor Attack Imperceptible Trigger

January 20, 2024

Detecting Hidden Triggers: Mapping Non-Markov Reward Functions to Markov
Gregory Hyde, Eugene Santos Jr
Reinforcement Learning Algorithm Markov Model Non Markovian Reward Imperceptible Trigger Markovian Reward

August 31, 2023

FTA: Stealthy and Adaptive Backdoor Attack with Flexible Triggers on Federated Learning
Yanqi Qiao, Dazhuang Liu, Congwen Chen, Rui Wang, Kaitai Liang
Digital STEALTH Metric Trigger Design Robust Backdoor Attack Imperceptible Trigger

July 17, 2023

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound
Hanbo Cai, Pengcheng Zhang, Hai Dong, Yan Xiao, Stefanos Koffas, Yiming Li
Speech Recognition Backdoor Attack Sound Design Backdoor Poisoning Attack Element Level Information Stealthy Backdoor Attack Imperceptible Trigger Inaudible Voice

June 2, 2023

Poisoning Network Flow Classifiers
Giorgio Severi, Simona Boboila, Alina Oprea, John Holodnak, Kendra Kratkiewicz, Jason Matterer
Adversarial Attack Backdoor Attack Traffic Classification Imperceptible Trigger

May 11, 2023

Backdoor Attack with Sparse and Invisible Trigger
Yinghua Gao, Yiming Li, Xueluan Gong, Zhifeng Li, Shu-Tao Xia, Qian Wang
Deep Neural Network Backdoor Attack Many Sparse Backdoor Defense Invisible Backdoor Attack Imperceptible Trigger

August 19, 2022

Dispersed Pixel Perturbation-based Imperceptible Backdoor Trigger for Image Classifier Models
Yulong Wang, Minghui Zhao, Shenghong Li, Xin Yuan, Wei Ni
Backdoor Attack Pixel Wise Backdoor Trigger Hidden Backdoor Imperceptible Trigger