v1v2 (latest)

Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks

International Conference on Learning Representations (ICLR), 2021

15 January 2021

ArXiv (abs)PDF HTML Github (122★)

Papers citing "Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks"

50 / 282 papers shown

Title
Backdoor Vulnerabilities in Normally Trained Deep Learning Models Guanhong Tao Zhenting Wang Shuyang Cheng Shiqing Ma Shengwei An Yingqi Liu Guangyu Shen Zhuo Zhang Yunshu Mao Xiangyu Zhang SILM 183 18 0 29 Nov 2022
BadPrompt: Backdoor Attacks on Continuous PromptsNeural Information Processing Systems (NeurIPS), 2022 Xiangrui Cai Haidong Xu Sihan Xu Ying Zhang Xiaojie Yuan SILM 198 82 0 27 Nov 2022
Backdoor Cleansing with Unlabeled DataComputer Vision and Pattern Recognition (CVPR), 2022 Lu Pang Tao Sun Haibin Ling Chao Chen AAML 267 27 0 22 Nov 2022
Invisible Backdoor Attack with Dynamic Triggers against Person Re-identificationIEEE Transactions on Information Forensics and Security (IEEE TIFS), 2022 Wenli Sun Xinyang Jiang Shuguang Dou Dongsheng Li Duoqian Miao Cheng Deng Cairong Zhao AAML 217 16 0 20 Nov 2022
Provable Defense against Backdoor Policies in Reinforcement LearningNeural Information Processing Systems (NeurIPS), 2022 S. Bharti Xuezhou Zhang Adish Singla Xiaojin Zhu AAML 155 26 0 18 Nov 2022
Backdoor Attacks on Time Series: A Generative Approach Yujing Jiang Jiabo He S. Erfani James Bailey AAML AI4TS 273 19 0 15 Nov 2022
M-to-N Backdoor Paradigm: A Multi-Trigger and Multi-Target Attack to Deep Learning Models Linshan Hou Zhongyun Hua Yuhong Li Yifeng Zheng Leo Yu Zhang AAML 275 6 0 03 Nov 2022
Dormant Neural TrojansInternational Conference on Machine Learning and Applications (ICMLA), 2022 Feisi Fu Panagiota Kiourti Wenchao Li AAML 205 0 0 02 Nov 2022
BATT: Backdoor Attack with Transformation-based TriggersIEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022 Tong Xu Yiming Li Yong Jiang Shutao Xia AAML 220 25 0 02 Nov 2022
Backdoor Defense via Suppressing Model ShortcutsIEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022 Shengyuan Yang Yiming Li Yong Jiang Shutao Xia AAML 194 14 0 02 Nov 2022
The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised LearningIEEE International Conference on Computer Vision (ICCV), 2022 Virat Shejwalkar Lingjuan Lyu Amir Houmansadr AAML 188 14 0 01 Nov 2022
Rethinking the Reverse-engineering of Trojan TriggersNeural Information Processing Systems (NeurIPS), 2022 Zhenting Wang Kai Mei Hailun Ding Juan Zhai Shiqing Ma 164 50 0 27 Oct 2022
Training set cleansing of backdoor poisoning by self-supervised representation learningIEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022 H. Wang Soroush Karami Ousmane Amadou Dia H. Ritter E. Emamjomeh-Zadeh J. Chen Zhen Xiang D. J. Miller G. Kesidis SSL 159 4 0 19 Oct 2022
Fine-mixing: Mitigating Backdoors in Fine-tuned Language ModelsConference on Empirical Methods in Natural Language Processing (EMNLP), 2022 Zhiyuan Zhang Lingjuan Lyu Jiabo He Chenguang Wang Xu Sun AAML 176 57 0 18 Oct 2022
Marksman Backdoor: Backdoor Attacks with Arbitrary Target ClassNeural Information Processing Systems (NeurIPS), 2022 Khoa D. Doan Yingjie Lao Ping Li 207 51 0 17 Oct 2022
Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor AttacksConference on Empirical Methods in Natural Language Processing (EMNLP), 2022 Sishuo Chen Wenkai Yang Zhiyuan Zhang Xiaohan Bi Xu Sun SILM AAML 177 32 0 14 Oct 2022
Dim-Krum: Backdoor-Resistant Federated Learning for NLP with Dimension-wise Krum-Based AggregationConference on Empirical Methods in Natural Language Processing (EMNLP), 2022 Zhiyuan Zhang Qi Su Xu Sun FedML 135 18 0 13 Oct 2022
COLLIDER: A Robust Training Framework for Backdoor DataAsian Conference on Computer Vision (ACCV), 2022 H. M. Dolatabadi S. Erfani C. Leckie AAML 179 8 0 13 Oct 2022
Trap and Replace: Defending Backdoor Attacks by Trapping Them into an Easy-to-Replace SubnetworkNeural Information Processing Systems (NeurIPS), 2022 Haotao Wang Junyuan Hong Aston Zhang Jiayu Zhou Zinan Lin AAML 199 16 0 12 Oct 2022
Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright ProtectionNeural Information Processing Systems (NeurIPS), 2022 Yiming Li Yang Bai Yong Jiang Yong-Liang Yang Shutao Xia Bo Li AAML 380 136 0 27 Sep 2022
Exploring Inconsistent Knowledge Distillation for Object Detection with Data AugmentationACM Multimedia (ACM MM), 2022 Jiawei Liang Yaning Tan Aishan Liu Ke Ma Jingzhi Li Xiaochun Cao VLM 306 18 0 20 Sep 2022
Confidence Matters: Inspecting Backdoors in Deep Neural Networks via Distribution Transfer Tong Wang Xingtai Lv F. Xu Miao Xu Shengwei An Ting Wang AAML 137 3 0 13 Aug 2022
Defense against Backdoor Attacks via Identifying and Purifying Bad Neurons Mingyuan Fan Yang Liu Cen Chen Ximeng Liu Wenzhong Guo AAML 126 4 0 13 Aug 2022
Data-free Backdoor Removal based on Channel LipschitznessEuropean Conference on Computer Vision (ECCV), 2022 Runkai Zheng Rong Tang Jianze Li Li Liu AAML 262 124 0 05 Aug 2022
Just Rotate it: Deploying Backdoor Attacks via Rotation Transformation Tong Wu Tianhao Wang Vikash Sehwag Saeed Mahloujifar Prateek Mittal AAML 176 45 0 22 Jul 2022
Game of Trojans: A Submodular Byzantine Approach D. Sahabandu Arezoo Rajabi Luyao Niu Yangqiu Song Bhaskar Ramasubramanian Radha Poovendran AAML 140 0 0 13 Jul 2022
Backdoor Attacks on Crowd CountingACM Multimedia (ACM MM), 2022 Yuhua Sun Tailai Zhang Jiabo He Pan Zhou Jian Lou Zichuan Xu Xing Di Yu Cheng Lichao AAML 199 16 0 12 Jul 2022
One-shot Neural Backdoor Erasing via Adversarial Weight MaskingNeural Information Processing Systems (NeurIPS), 2022 Shuwen Chai Jinghui Chen AAML 226 43 0 10 Jul 2022
Defense Against Multi-target Trojan Attacks Haripriya Harikumar Santu Rana Kien Do Sunil R. Gupta W. Zong Willy Susilo Svetha Venkatesh AAML 140 4 0 08 Jul 2022
BackdoorBench: A Comprehensive Benchmark of Backdoor LearningNeural Information Processing Systems (NeurIPS), 2022 Baoyuan Wu Hongrui Chen Ruotong Wang Zihao Zhu Shaokui Wei Danni Yuan Chaoxiao Shen ELM AAML 266 183 0 25 Jun 2022
Natural Backdoor Datasets Emily Wenger Roma Bhattacharjee A. Bhagoji Josephine Passananti Emilio Andere Haitao Zheng Ben Y. Zhao AAML 197 6 0 21 Jun 2022
DECK: Model Hardening for Defending Pervasive Backdoors Guanhong Tao Yingqi Liu Shuyang Cheng Shengwei An Zhuo Zhang Qiuling Xu Guangyu Shen Xiangyu Zhang AAML 334 7 0 18 Jun 2022
Turning a Curse into a Blessing: Enabling In-Distribution-Data-Free Backdoor Removal via Stabilized Model Inversion Si-An Chen Yi Zeng J. T.Wang Won Park Xun Chen Lingjuan Lyu Zhuoqing Mao R. Jia 150 3 0 14 Jun 2022
BadDet: Backdoor Attacks on Object Detection Shih-Han Chan Yinpeng Dong Junyi Zhu Xiaolu Zhang Jun Zhou AAML 129 72 0 28 May 2022
Towards A Proactive ML Approach for Detecting Backdoor Poison SamplesUSENIX Security Symposium (USENIX Security), 2022 Xiangyu Qi Tinghao Xie Jiachen T. Wang Tong Wu Saeed Mahloujifar Prateek Mittal AAML 291 68 0 26 May 2022
BppAttack: Stealthy and Efficient Trojan Attacks against Deep Neural Networks via Image Quantization and Contrastive Adversarial LearningComputer Vision and Pattern Recognition (CVPR), 2022 Zhenting Wang Juan Zhai Shiqing Ma AAML 273 134 0 26 May 2022
BITE: Textual Backdoor Attacks with Iterative Trigger InjectionAnnual Meeting of the Association for Computational Linguistics (ACL), 2022 Jun Yan Vansh Gupta Xiang Ren SILM 411 73 0 25 May 2022
Quarantine: Sparsity Can Uncover the Trojan Attack Trigger for FreeComputer Vision and Pattern Recognition (CVPR), 2022 Tianlong Chen Zhenyu Zhang Yihua Zhang Shiyu Chang Sijia Liu Zinan Lin AAML 165 27 0 24 May 2022
Towards a Defense Against Federated Backdoor Attacks Under Continuous Training Shuai Wang J. Hayase Giulia Fanti Sewoong Oh FedML 273 7 0 24 May 2022
MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statistic Hang Wang Zhen Xiang David J. Miller G. Kesidis AAML 271 60 0 13 May 2022
Model-Contrastive Learning for Backdoor Defense Zhihao Yue Jun Xia Zhiwei Ling Ming Hu Ting Wang Xian Wei Xiao He AAML 140 3 0 09 May 2022
Imperceptible Backdoor Attack: From Input Space to Feature RepresentationInternational Joint Conference on Artificial Intelligence (IJCAI), 2022 Nan Zhong Zhenxing Qian Xinpeng Zhang AAML 133 64 0 06 May 2022
Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data PoisoningACM Computing Surveys (ACM CSUR), 2022 Antonio Emanuele Cinà Kathrin Grosse Ambra Demontis Sebastiano Vascon Werner Zellinger Bernhard A. Moser Alina Oprea Battista Biggio Marcello Pelillo Fabio Roli AAML 367 166 0 04 May 2022
Eliminating Backdoor Triggers for Deep Neural Networks Using Attention Relation Graph DistillationInternational Joint Conference on Artificial Intelligence (IJCAI), 2022 Jun Xia Ting Wang Jiepin Ding Xian Wei Xiao He AAML 185 37 0 21 Apr 2022
Robust Unlearnable Examples: Protecting Data Against Adversarial Learning Shaopeng Fu Fengxiang He Yang Liu Li Shen Dacheng Tao 145 36 0 28 Mar 2022
Trojan Horse Training for Breaking Defenses against Backdoor Attacks in Deep Learning Arezoo Rajabi Bhaskar Ramasubramanian Radha Poovendran AAML 200 5 0 25 Mar 2022
Towards Effective and Robust Neural Trojan Defenses via Input FilteringEuropean Conference on Computer Vision (ECCV), 2022 Kien Do Haripriya Harikumar Hung Le D. Nguyen T. Tran Santu Rana Dang Nguyen Willy Susilo Svetha Venkatesh AAML 200 13 0 24 Feb 2022
On the Effectiveness of Adversarial Training against Backdoor AttacksIEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2022 Yinghua Gao Dongxian Wu Jingfeng Zhang Guanhao Gan Shutao Xia Gang Niu Masashi Sugiyama AAML 169 29 0 22 Feb 2022
Label-Smoothed Backdoor Attack Minlong Peng Zidi Xiong Mingming Sun Ping Li AAML 150 4 0 19 Feb 2022
A Survey of Neural Trojan Attacks and Defenses in Deep Learning Jie Wang Ghulam Mubashar Hassan Naveed Akhtar AAML 178 27 0 15 Feb 2022