mlsanityData to 14 Oct 2026

TED-LaST: Towards Robust Backdoor Defense Against Adaptive Attacks

arXiv 2506.10722 ↗First posted 12 Jun 2025cs.CR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep Neural Networks (DNNs) are vulnerable to backdoor attacks, where attackers implant hidden triggers during training to maliciously control model behavior. Topological Evolution Dynamics (TED) has recently emerged as a powerful tool for detecting backdoor attacks in DNNs. However, TED can be vulnerable to backdoor attacks that adaptively distort topological representation distributions across network layers. To address this limitation, we propose TED-LaST (Topological Evolution Dynamics against Laundry, Slow release, and Target mapping attack strategies), a novel defense strategy that enhances TED's robustness against adaptive attacks. TED-LaST introduces two key innovations: label-supervised dynamics tracking and adaptive layer emphasis. These enhancements enable the identification of stealthy threats that evade traditional TED-based defenses, even in cases of inseparability in topological space and subtle topological perturbations. We review and classify data poisoning tricks in state-of-the-art adaptive attacks and propose enhanced adaptive attack with target mapping, which can dynamically shift malicious tasks and fully leverage the stealthiness that adaptive attacks possess. Our comprehensive experiments on multiple datasets (CIFAR-10, GTSRB, and ImageNet100) and model architectures (ResNet20, ResNet101) show that TED-LaST effectively counteracts sophisticated backdoors like Adap-Blend, Adapt-Patch, and the proposed enhanced adaptive attack. TED-LaST sets a new benchmark for robust backdoor detection, substantially enhancing DNN security against evolving threats.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 31 of its 52 references that are in the map. First posted on arXiv 12 Jun 2025.

Builds on

31 of 52 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  3. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  4. 2019Deep Learning for 3D Point Clouds: A Surveycited by 558
  5. 2018Spectral Signatures in Backdoor Attackscited by 484
  6. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
  7. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  8. 2018Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooringcited by 357
  9. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 353
  10. 2018Deep Audio-Visual Speech Recognitioncited by 322
  11. 2020Input-Aware Dynamic Backdoor Attackcited by 313
  12. 2019PyOD: A Python Toolbox for Scalable Outlier Detectioncited by 222
19 more
  1. 2019Detecting AI Trojans Using Meta Neural Analysiscited by 199
  2. 2020Blind Backdoors in Deep Learning Modelscited by 182
  3. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspectivecited by 161
  4. 2022Backdoor Defense via Decoupling the Training Processcited by 153
  5. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 130
  6. 2019Demon in the Variant: Statistical Analysis of DNNs for Robust Backdoor Contamination Detectioncited by 124
  7. 2021Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoningcited by 113
  8. 2021Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Modelscited by 105
  9. 2020An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networkscited by 97
  10. 2019Bypassing Backdoor Detection Algorithms in Deep Learningcited by 74
  11. 2018Backdooring Convolutional Neural Networks via Targeted Weight Perturbationscited by 69
  12. 2023Detecting Backdoors During the Inference Stage Based on Corruption Robustness Consistencycited by 46
  13. 2020From Handcrafted to Deep Features for Pedestrian Detection: A Surveycited by 30
  14. 2020Backdooring and Poisoning Neural Networks with Image-Scaling Attackscited by 28
  15. 2023Robust Backdoor Detection for Deep Learning via Topological Evolution Dynamicscited by 26
  16. 2023Defending Against Backdoor Attacks by Layer-wise Feature Analysiscited by 16
  17. 2021Poisoning Attack against Estimating from Pairwise Comparisonscited by 14
  18. 2020Odyssey: Creation, Analysis and Detection of Trojan Modelscited by 7
  19. 2023Universal Detection of Backdoor Attacks via Density-based Clustering and Centroids Analysiscited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025TED++: Submanifold-Aware Backdoor Detection via Layerwise Tubular-Neighbourhood Screening
  2. 2026Learning Normal Diffusion Dynamics for Backdoor Defense in Text-to-Image Models
  3. 2026Trapping Attacker in Dilemma: Examining Internal Correlations and External Influences of Trigger for Defending GNN Backdoors
  4. 2025Defending against Backdoor Attacks via Module Switching
  5. 2023BELT: Old-School Backdoor Attacks can Evade the State-of-the-Art Defense with Backdoor Exclusivity Lifting
  6. 2024Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack
  7. 2024Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
  8. 2021Dyn-Backdoor: Backdoor Attack on Dynamic Link Prediction