mlsanityData to 14 Oct 2026

Shortcuts Everywhere and Nowhere: Exploring Multi-Trigger Backdoor Attacks

arXiv 2401.15295 ↗First posted 27 Jan 2024cs.LG cs.CRcited by 19

Not on the map. Why?

Abstract (arXiv)

Backdoor attacks have become a significant threat to the pre-training and deployment of deep neural networks (DNNs). Although numerous methods for detecting and mitigating backdoor attacks have been proposed, most rely on identifying and eliminating the ``shortcut" created by the backdoor, which links a specific source class to a target class. However, these approaches can be easily circumvented by designing multiple backdoor triggers that create shortcuts everywhere and therefore nowhere specific. In this study, we explore the concept of Multi-Trigger Backdoor Attacks (MTBAs), where multiple adversaries leverage different types of triggers to poison the same dataset. By proposing and investigating three types of multi-trigger attacks including textitparallel, textitsequential, and textithybrid attacks, we demonstrate that 1) multiple triggers can coexist, overwrite, or cross-activate one another, and 2) MTBAs easily break the prevalent shortcut assumption underlying most existing backdoor detection/removal methods, rendering them ineffective. Given the security risk posed by MTBAs, we have created a multi-trigger backdoor poisoning dataset to facilitate future research on detecting and mitigating these attacks, and we also discuss potential defense strategies against MTBAs. Our code is available at https://github.com/bboylyg/Multi-Trigger-Backdoor-Attacks.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Builds on 33 of its 49 references that are in the map. First posted on arXiv 27 Jan 2024.

Builds on

33 of 49 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  6. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  7. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  8. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 580
  9. 2015Listen, Attend and Spellcited by 543
  10. 2018Spectral Signatures in Backdoor Attackscited by 484
  11. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
  12. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
21 more
  1. 2020Backdoor Learning: A Surveycited by 426
  2. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 353
  3. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 349
  4. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 324
  5. 2020Input-Aware Dynamic Backdoor Attackcited by 313
  6. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 263
  7. 2019A new Backdoor Attack in CNNs by training set corruption without label poisoningcited by 242
  8. 2021Adversarial Neuron Pruning Purifies Backdoored Deep Modelscited by 235
  9. 2019Detecting AI Trojans Using Meta Neural Analysiscited by 199
  10. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspectivecited by 161
  11. 2022Backdoor Defense via Decoupling the Training Processcited by 153
  12. 2019TABOR: A Highly Accurate Approach to Inspecting and Restoring Trojan Backdoors in AI Systemscited by 122
  13. 2020Deep Feature Space Trojan Attack of Neural Networks by Controlled Detoxificationcited by 95
  14. 2022Data-free Backdoor Removal based on Channel Lipschitznesscited by 79
  15. 2020Rethinking the Trigger of Backdoor Attackcited by 77
  16. 2023Reconstructive Neuron Pruning for Backdoor Defensecited by 59
  17. 2022MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statisticcited by 55
  18. 2023Distilling Cognitive Backdoor Patterns within an Imagecited by 20
  19. 2023UMD: Unsupervised Model Detection for X2X Backdoor Attackscited by 18
  20. 2022Imperceptible and Multi-channel Backdoor Attack against Deep Neural Networkscited by 10
  21. 2022M-to-N Backdoor Paradigm: A Multi-Trigger and Multi-Target Attack to Deep Learning Modelscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 11; 2026: 6.

19 papers.

Sort:

By area:

  1. 2025Safety at Scale: A Comprehensive Survey of Large Model and Agent Safetycited by 67
  2. 2024CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularizationcited by 33
  3. 2026Purifying Generative LLMs from Backdoors without Prior Knowledge or Clean Referencecited by 7
  4. 2025P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMscited by 6
  5. 2025Pay Attention to the Triggers: Constructing Backdoors That Survive Distillationcited by 6
  6. 2025Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMscited by 4
  7. 2025BADTV: Unveiling Backdoor Threats in Third-Party Task Vectorscited by 3
  8. 2024Non-Cooperative Backdoor Attacks in Federated Learning: A New Threat Landscapecited by 1
  9. 2025DISTIL: Data-Free Inversion of Suspicious Trojan Inputs via Latent Diffusioncited by 1
  10. 2026Kill it with FIRE: On Leveraging Latent Space Directions for Runtime Backdoor Mitigation in Deep Neural Networkscited by 1
  11. 2025Defending against Backdoor Attacks via Module Switchingcited by 0
  12. 2025Proactive Disentangled Modeling of Trigger-Object Pairings for Backdoor Defensecited by 0
7 more
  1. 2025Backdoor Samples Detection Based on Perturbation Discrepancy Consistency in Pre-trained Language Modelscited by 0
  2. 2025Localizing Malicious Outputs from CodeLLMcited by 0
  3. 2025Microsaccade-Inspired Probing: Positional Encoding Perturbations Reveal LLM Misbehaviourscited by 0
  4. 2026Cordyceps: Covert Control Attacks on LLMs via Data Poisoningcited by 0
  5. 2026LoRAScan: Detecting Backdoor Prompts in Low-Rank Adapters for Large Language Models via Down-Projection Activation Spikescited by 0
  6. 2026Backdoor Containment via Expert Quarantine and Shutdown in LLMscited by 0
  7. 2026Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projectioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025A4O: All Trigger for One sample
  2. 2024A Practical Trigger-Free Backdoor Attack on Neural Networks ↗
  3. 2025Isolate Trigger: Detecting and Eliminating Adaptive Backdoor Attacks ↗
  4. 2022Imperceptible Backdoor Attack: From Input Space to Feature Representation
  5. 2025DeBackdoor: A Deductive Framework for Detecting Backdoor Attacks on Deep Models with Limited Data
  6. 2023Backdoor Attack with Sparse and Invisible Trigger
  7. 2022BackdoorBench: A Comprehensive Benchmark of Backdoor Learning
  8. 2023Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective