Research map / Adversarial robustness and security / Backdoor and poisoning
Flatness-aware Sequential Learning Generates Resilient Backdoors
Hoang Pham, The-Anh Ta, Anh T Tran, Khoa D Doan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train backdoor with SBL (two-task sequential learning with SAM and CL) instead of conventional training | BadNets, Blended, SIG, Dynamic · CIFAR-10 / GTSRB, fine-tuning defenses | ASR after defense: CIFAR-10 BadNets FT SGD-0.01 ASR 1.79 -> 100 | Improved |
| train backdoor with SBL | BadNets on ImageNet-10 · ImageNet-10 after FT SGD-0.01 | ASR and clean accuracy: ASR 10.47 -> 76.54; CA 51.38 -> 85.19 | Improved |
| add continual learning regularization (Naive, EWC, Anchoring, AGEM) in step 1 without SAM (ablation) | SBL on BadNets · CIFAR-10 FT SGD-0.01 | ASR: 1.79 -> 23.19 (Naive), 85.73 (Anchoring) | Improved |
| add continual learning in step 1 without SAM (ablation) | SBL on BadNets · CIFAR-10 FT-SAM | ASR: 2.5 -> 2.84 (Naive), 13.6 (AGEM) | No clear effect |
| add SAM to conventional backdoor training (ablation) | Conventional backdoor learning on BadNets · CIFAR-10 fine-tuning defenses | ASR: FT-SAM 2.5 -> 99.99; FT SGD-0.01 1.79 -> 3.14 | Mixed |
| add SAM in step 0 of SBL w. EWC (ablation) | SBL w. EWC · CIFAR-10 BadNets | ASR after FT-SAM: 6.13 -> 100 | Improved |
| use large step-1 learning rate 0.01 for naive CL (ablation) | SBL w. Naive · CIFAR-10 BadNets | ASR after FT SGD-0.01: 100 -> 6.34 | Hurt |
| use EWC at large step-1 learning rate 0.01 (ablation) | SBL · CIFAR-10 BadNets | ASR and CA: ASR 79.91 retained, step 1 CA 85.32 | Mixed |
| lower the poisoning rate to 1% (ablation) | BadNets w. SBL · CIFAR-10 | ASR after FT-SAM: 99.80 (vs 4.01 for plain BadNets) | No clear effect |
| apply SBL to other architectures (ablation) | BadNets, Blended, SIG · CIFAR-10 ResNet-20 and VGG-16 | ASR after defenses: VGG-16 BadNets FT SGD-0.01 3.20 -> 100 | Improved |
Closest papers by meaning
- Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness (NeurIPS 2024 · similarity 0.90)
- Attack on Prompt: Backdoor Attack in Prompt-Based Continual Learning (AAAI 2025 · similarity 0.89)
- BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model (NeurIPS 2025 · similarity 0.89)
- Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense (NeurIPS 2024 · similarity 0.88)
- Backdoor Mitigation via Invertible Pruning Masks (NeurIPS 2025 · similarity 0.87)
- Demystifying Poisoning Backdoor Attacks from a Statistical Perspective (ICLR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
