DeBackdoor: A Deductive Framework for Detecting Backdoor Attacks on Deep Models with Limited Data

arXiv 2503.21305 ↗First posted 27 Mar 2025cs.CR cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Backdoor attacks are among the most effective, practical, and stealthy attacks in deep learning. In this paper, we consider a practical scenario where a developer obtains a deep model from a third party and uses it as part of a safety-critical system. The developer wants to inspect the model for potential backdoors prior to system deployment. We find that most existing detection techniques make assumptions that are not applicable to this scenario. In this paper, we present a novel framework for detecting backdoors under realistic restrictions. We generate candidate triggers by deductively searching over the space of possible triggers. We construct and optimize a smoothed version of Attack Success Rate as our search objective. Starting from a broad class of template attacks and just using the forward pass of a deep model, we reverse engineer the backdoor attack. We conduct extensive evaluation on a wide range of attacks, models, and datasets, with our technique performing almost perfectly across these settings.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 53 of its 82 references that are in the map. First posted on arXiv 27 Mar 2025.

Builds on

53 of 82 references.

References · 53

2015Deep Residual Learning for Image Recognitioncited by 52,096
2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
2014Going Deeper with Convolutionscited by 4,241
2016Wide Residual Networkscited by 3,070
2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
2016End to End Learning for Self-Driving Carscited by 1,009
2018How To Backdoor Federated Learningcited by 765
2018Spectral Signatures in Backdoor Attackscited by 484
2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
2020Backdoor Learning: A Surveycited by 426

38 more

2019Hidden Trigger Backdoor Attackscited by 371
2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 349
2019Label-Consistent Backdoor Attackscited by 296
2019A new Backdoor Attack in CNNs by training set corruption without label poisoningcited by 242
2021Adversarial Neuron Pruning Purifies Backdoored Deep Modelscited by 235
2019A backdoor attack against LSTM-based text classification systemscited by 227
2019Detecting AI Trojans Using Meta Neural Analysiscited by 199
2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 152
2020Dynamic Backdoor Attacks Against Machine Learning Modelscited by 151
2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 146
2020Backdoor Attacks Against Deep Learning Systems in the Physical Worldcited by 134
2020Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustnesscited by 132
2020Backdoor Attacks to Graph Neural Networkscited by 128
2021Backdoor Scanning for Deep Neural Networks through K-Arm Optimizationcited by 98
2022Better plain ViT baselines for ImageNet-1kcited by 98
2023SCALE-UP: An Efficient Black-box Input-level Backdoor Detection via Analyzing Scaled Prediction Consistencycited by 97
2020Practical Detection of Trojan Neural Networks: Data-Limited and Data-Free Casescited by 95
2020Rethinking the Trigger of Backdoor Attackcited by 77
2021Black-box Detection of Backdoor Attacks with Limited Information and Datacited by 73
2018Backdooring Convolutional Neural Networks via Targeted Weight Perturbationscited by 69
2024Instruction Backdoor Attacks Against Customized LLMscited by 62
2021AEVA: Black-box Backdoor Detection Using Adversarial Extreme Value Analysiscited by 53
2019Model Agnostic Defence against Backdoor Attacks in Machine Learningcited by 52
2020Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Modelscited by 44
2018Metropolis-Hastings Generative Adversarial Networkscited by 41
2022Towards A Proactive ML Approach for Detecting Backdoor Poison Samplescited by 40
2024An LLM-Assisted Easy-to-Trigger Backdoor Attack on Code Completion Models: Injecting Disguised Vulnerabilities against Strong Detectioncited by 40
2020Stop-and-Go: Exploring Backdoor Attacks on Deep Reinforcement Learning-based Traffic Congestion Control Systemscited by 37
2021Topological Detection of Trojaned Neural Networkscited by 34
2023ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning Paradigmscited by 29
2020Don't Trigger Me! A Triggerless Backdoor Attack Against Deep Neural Networkscited by 19
2024BackdoorIndicator: Leveraging OOD Data for Proactive Backdoor Detection in Federated Learningcited by 17
2021Fall of Giants: How popular text-based MLaaS fall against a simple evasion attackcited by 13
2022Backdoor Vulnerabilities in Normally Trained Deep Learning Modelscited by 13
2020NNoculation: Catching BadNets in the Wildcited by 9
2023On the Difficulty of Defending Contrastive Learning against Backdoor Attackscited by 9
2020Towards Class-Oriented Poisoning Attacks Against Neural Networkscited by 8
2023Look, Listen, and Attack: Backdoor Attacks Against Video Action Recognitioncited by 7

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

2026Beyond Crash: Hijacking Your Autonomous Vehicle for Fun and Profitcited by 1
2026Beyond Attack Success Rate: A Multi-Metric Evaluation of Adversarial Transferability in Medical Imaging Modelscited by 1
2026Checkerboard: Closed-Form and Data-Independent Trigger Design for Clean-Label Backdoor Attackscited by 0
2026Fast and Lightweight Backdoor Detection via Head Random Probingcited by 0
2026Detecting Backdoors in Object Detection via Pre-NMS Prediction Distribution Shiftcited by 0

Similar, not cited

Close in wording, no citation either way