Backdoor Learning

Backdoor learning describes the malicious insertion of vulnerabilities into machine learning models during training, causing them to produce incorrect outputs when presented with specific "trigger" inputs. Current research focuses on developing both more sophisticated attacks and robust defenses across various model architectures, including deep neural networks (DNNs), graph neural networks (GNNs), and multimodal contrastive learning models like CLIP, often employing techniques like unlearning, model pruning, and contrastive learning. Understanding and mitigating backdoor attacks is crucial for ensuring the trustworthiness and security of machine learning systems deployed in sensitive applications, driving ongoing efforts to establish standardized benchmarks and develop effective detection and mitigation strategies.

Papers

November 17, 2024

BackdoorMBTI: A Backdoor Learning Multimodal Benchmark Tool Kit for Backdoor Defense Evaluation
Haiyang Yu, Tian Xie, Jiaping Gui, Pengyang Wang, Ping Yi, Yue Wu
Backdoor Attack Backdoor Defense Backdoor Learning Mitigating Backdoor Backdoor Model Multimodal Benchmark Datasets

October 18, 2024

Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation
Shuai Zhao, Xiaobao Wu, Cong-Duy Nguyen, Meihuizi Jia, Yichao Feng, Luu Anh Tuan
Language Model Medical LLM Backdoor Attack Mutual Distillation Backdoor Learning Weak to Strong

October 14, 2024

How to Backdoor Consistency Models?
Chengen Wang, Murat Kantarcioglu
Adversarial Attack Backdoor Attack Consistency Model Backdoor Learning Backdoor Model Image Backdoor Attack

October 2, 2024

"No Matter What You Do": Purifying GNN Models via Backdoor Unlearning
Jiale Zhang, Chengcheng Zhu, Bosen Rao, Hao Sui, Xiaobing Sun, Bing Chen, Chunyi Zhou, Shouling Ji
Graph Neural Network Backdoor Attack Backdoor Defense Backdoor Learning Backdoor Removal

September 29, 2024

Efficient Backdoor Defense in Multimodal Contrastive Learning: A Token-Level Unlearning Method for Mitigating Threats
Kuanrong Liu, Siyuan Liang, Jiawei Liang, Pengwen Dai, Xiaochun Cao
Contrastive Learning Backdoor Learning Backdoor Effect Input Level Backdoor Detection Backdoor Policy

July 29, 2024

BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning
Baoyuan Wu, Hongrui Chen, Mingda Zhang, Zihao Zhu, Shaokui Wei, Danni Yuan, Mingli Zhu, Ruotong Wang, Li Liu, Chao Shen
General Analysis Comprehensive Benchmark Backdoor Learning

July 20, 2024

Flatness-aware Sequential Learning Generates Resilient Backdoors
Hoang Pham, The-Anh Ta, Anh Tran, Khoa D. Doan
Fine Tuning Backdoor Attack Sequential Learning Backdoor Learning Input Level Backdoor Detection Resilient Backdoor

May 7, 2024

Unlearning Backdoor Attacks through Gradient-Based Model Pruning
Kealan Dunnett, Reza Arablouei, Dimity Miller, Volkan Dedeoglu, Raja Jurdak
Convolutional Neural Network Backdoor Attack Backdoor Learning Loss Gradient

March 26, 2024

Identifying Backdoored Graphs in Graph Neural Network Training: An Explanation-Based Approach with Novel Metrics
Jane Downer, Ren Wang, Binghui Wang
Graph Neural Network Backdoor Attack Person Identification Backdoor Trigger GNN Explanation Backdoor Learning Graph Explainability

March 24, 2024

Unlearning Backdoor Threats: Enhancing Backdoor Defense in Multimodal Contrastive Learning via Local Token Unlearning
Siyuan Liang, Kuanrong Liu, Jiajun Gong, Jiawei Liang, Yuan Xun, Ee-Chien Chang, Xiaochun Cao
Contrastive Learning Backdoor Attack Token Level Backdoor Defense Backdoor Learning Training Pair

February 26, 2024

On the (In)feasibility of ML Backdoor Detection as an Hypothesis Testing Problem
Georg Pichler, Marco Romanelli, Divya Prakash Manivannan, Prashanth Krishnamurthy, Farshad Khorrami, Siddharth Garg
Distribution Detection Hypothesis Testing Backdoor Detection Backdoor Learning Feasibility Guarantee

February 19, 2024

Acquiring Clean Language Models from Backdoor Poisoned Datasets by Downscaling Frequency Space
Zongru Wu, Zhuosheng Zhang, Pengzhou Cheng, Gongshen Liu
Language Model Statistical Downscaling Backdoor Learning Backdoor Data

January 26, 2024

BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning
Baoyuan Wu, Hongrui Chen, Mingda Zhang, Zihao Zhu, Shaokui Wei, Danni Yuan, Mingli Zhu, Ruotong Wang, Li Liu, Chao Shen
General Analysis Comprehensive Benchmark Backdoor Learning

January 19, 2024

Learning Backdoors for Mixed Integer Programs with Contrastive Learning
Junyang Cai, Taoan Huang, Bistra Dilkina
Contrastive Learning Backdoor Attack Monte Carlo Tree Search Mixed Integer Backdoor Learning

January 6, 2024

End-to-End Anti-Backdoor Learning on Images and Time Series
Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani, Yige Li, James Bailey
Deep Learning Model Time Series Backdoor Attack Backdoor Trigger Backdoor Learning

December 13, 2023

Erasing Self-Supervised Learning Backdoor by Cluster Activation Masking
Shengsheng Qian, Dizhan Xue, Yifei Wang, Shengjie Zhang, Huaiwen Zhang, Changsheng Xu
Backdoor Attack Backdoor Trigger Backdoor Learning

December 4, 2023

OCGEC: One-class Graph Embedding Classification for DNN Backdoor Detection
Haoyu Jiang, Haiyang Yu, Nan Li, Ping Yi
Classification Code Backdoor Learning Backdoor Sample Backdoor Model

November 20, 2023

BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
Siyuan Liang, Mingli Zhu, Aishan Liu, Baoyuan Wu, Xiaochun Cao, Ee-Chien Chang
Contrastive Learning Backdoor Attack Backdoor Defense Backdoor Detection Backdoor Learning

July 11, 2023

Differential Analysis of Triggers and Benign Features for Black-Box DNN Backdoor Detection
Hao Fu, Prashanth Krishnamurthy, Siddharth Garg, Farshad Khorrami
Deep Neural Network Backdoor Attack Novelty Detection Detection Method Backdoor Learning Natural Trigger Differential Analysis

June 13, 2023

Privacy Inference-Empowered Stealthy Backdoor Attack on Federated Learning under Non-IID Scenarios
Haochen Mei, Gaolei Li, Jun Wu, Longfei Zheng
Privacy Policy Multi Scenario Backdoor Learning Stealthy Backdoor Attack Privacy Inference