A survey of backdoor attacks and defences: From deep neural networks to large language models

Deep neural networks (DNNs) have found extensive applications in safety-critical artificial intelligence systems, such as autonomous driving and facial recognition systems. However, recent research has revealed their susceptibility to backdoors maliciously injected by adversaries. This vulnerability...

書誌詳細
出版年:	Journal of Electronic Science and Technology
主要な著者:	Ling-Xin Jin, Wei Jiang, Xiang-Yu Wen, Mei-Yu Lin, Jin-Yu Zhan, Xing-Zhi Zhou, Maregu Assefa Habtie, Naoufel Werghi
フォーマット:	論文
言語:	英語
出版事項:	KeAi Communications Co., Ltd. 2025-09-01
主題:	Backdoor attacks Backdoor defenses Deep neural networks Large language model
オンライン･アクセス:	http://www.sciencedirect.com/science/article/pii/S1674862X25000278

類似資料

Backdoor Pony: Evaluating backdoor attacks and defenses in different domains
著者:: Arthur Mercier, 等
出版事項: (2023-05-01)

Defending Against Backdoor Attacks by Quarantine Training
著者:: Chengxu Yu, 等
出版事項: (2024-01-01)

Defense Method Against Backdoor Attack in Federated Learning for Industrial Scenarios
著者:: WANG Xun, XU Fangmin, ZHAO Chenglin, LIU Hongfu
出版事項: (2024-01-01)

Sub-Band Backdoor Attack in Remote Sensing Imagery
著者:: Kazi Aminul Islam, 等
出版事項: (2024-04-01)

Efficient Method for Robust Backdoor Detection and Removal in Feature Space Using Clean Data
著者:: Donik Vrsnak, 等
出版事項: (2025-01-01)

An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defences
著者:: Wei Guo, 等
出版事項: (2022-01-01)

A4FL: Federated Adversarial Defense via Adversarial Training and Pruning Against Backdoor Attack
著者:: Saeed-Uz-Zaman, 等
出版事項: (2025-01-01)

Backdoor defense method in federated learning based on contrastive training
著者:: Jiale ZHANG, 等
出版事項: (2024-03-01)

Backdoor defense method in federated learning based on contrastive training
著者:: Jiale ZHANG, 等
出版事項: (2024-03-01)

Research on method and architecture for defense assessment of artificial intelligence backdoors
著者:: XIE Tian, 等
出版事項: (2024-09-01)

Survey on Backdoor Attacks and Countermeasures in Deep Neural Network
著者:: QIAN Hanwei, SUN Weisong
出版事項: (2023-05-01)

A Comprehensive Survey on Backdoor Attacks and Their Defenses in Face Recognition Systems
著者:: Quentin Le Roux, 等
出版事項: (2024-01-01)

DLP: towards active defense against backdoor attacks with decoupled learning process
著者:: Zonghao Ying, 等
出版事項: (2023-05-01)

A Novel Backdoor Detection Approach Using Entropy-Based Measures
著者:: Hema Karnam Surendrababu, 等
出版事項: (2024-01-01)

Exploring Graph Neural Backdoors in Vehicular Networks: Fundamentals, Methodologies, Applications, and Future Perspectives
著者:: Xiao Yang, 等
出版事項: (2025-01-01)

Survey on Backdoor Attacks on Deep Learning: Current Trends, Categorization, Applications, Research Challenges, and Future Prospects
著者:: Muhammad Abdullah Hanif, 等
出版事項: (2025-01-01)

Strategic safeguarding: A game theoretic approach for analyzing attacker-defender behavior in DNN backdoors
著者:: Kassem Kallas, 等
出版事項: (2024-10-01)

Textual Backdoor Defense via Poisoned Sample Recognition
著者:: Kun Shao, 等
出版事項: (2021-10-01)

Homophily-Guided Backdoor Attacks on GNN-Based Link Prediction
著者:: Yadong Wang, 等
出版事項: (2025-09-01)

Backdoor Attacks to Deep Learning Models and Countermeasures: A Survey
著者:: Yudong Li, 等
出版事項: (2023-01-01)

A Textual Backdoor Defense Method Based on Deep Feature Classification
著者:: Kun Shao, 等
出版事項: (2023-01-01)

XMAM:X-raying models with a matrix to reveal backdoor attacks for federated learning
著者:: Jianyi Zhang, 等
出版事項: (2024-08-01)

CLB-Defense: based on contrastive learning defense for graph neural network against backdoor attack
著者:: Jinyin CHEN, 等
出版事項: (2023-04-01)

CLB-Defense: based on contrastive learning defense for graph neural network against backdoor attack
著者:: Jinyin CHEN, 等
出版事項: (2023-04-01)

A Backdoor Approach With Inverted Labels Using Dirty Label-Flipping Attacks
著者:: Orson Mengara
出版事項: (2025-01-01)

Vulnerability analysis of federated learning malware detection systems based on backdoor attacks
著者:: LU Xingyu; CAO Yang
出版事項: (2024-09-01)

DriNet: Dynamic Backdoor Attack against Automatic Speech Recognization Models
著者:: Jianbin Ye, 等
出版事項: (2022-06-01)

Self-supervised Backdoor Attack Defence Method Based on Poisoned Classifier
著者:: WANG Yifei, ZHANG Shengjie, XUE Dizhan, QIAN Shengsheng
出版事項: (2025-04-01)

A sparse and invisible targeted backdoor attack in federated learning
著者:: Qikun Zhang, 等
出版事項: (2025-07-01)

Defending Deep Neural Networks Against Backdoor Attack by Using De-Trigger Autoencoder
著者:: Hyun Kwon
出版事項: (2025-01-01)

CheatKD:Knowledge Distillation Backdoor Attack Method Based on Poisoned Neuronal Assimilation
著者:: CHEN Jinyin, LI Xiao, JIN Haibo, CHEN Ruoxi, ZHENG Haibin, LI Hu
出版事項: (2024-03-01)

Camouflage Backdoor Attack against Pedestrian Detection
著者:: Yalun Wu, 等
出版事項: (2023-11-01)

A Backdoor Attack Against LSTM-Based Text Classification Systems
著者:: Jiazhu Dai, 等
出版事項: (2019-01-01)

DAGUARD: distributed backdoor attack defense scheme under federated learning
著者:: Shengxing YU, 等
出版事項: (2023-05-01)

DAGUARD: distributed backdoor attack defense scheme under federated learning
著者:: Shengxing YU, 等
出版事項: (2023-05-01)

EEG-Based Brain–Computer Interfaces are Vulnerable to Backdoor Attacks
著者:: Lubin Meng, 等
出版事項: (2023-01-01)

Clean-label backdoor attack on link prediction task
著者:: Junming Mo, 等
出版事項: (2025-08-01)

FLARE: A Backdoor Attack to Federated Learning with Refined Evasion
著者:: Qingya Wang, 等
出版事項: (2024-11-01)

Federated Learning Backdoor Attack Based on Frequency Domain Injection
著者:: Jiawang Liu, 等
出版事項: (2024-02-01)

Backdoor Attack on Deep Learning Models:A Survey
著者:: YING Zonghao, WU Bin
出版事項: (2023-03-01)