Research map / Adversarial robustness and security / Black-box and transfer attacks
Tracing the Origin of Adversarial Attack for Forensic Investigation and Deterrence
Han Fang, Jiyi Zhang, Yupeng Qiu, Jiayang Liu, Ke Xu, Chengfang Fang, Ee-Chien Chang
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 tracer with noise-sensitive loss instead of random initialization only (ablation) | tracer in separate-and-trace framework · CIFAR10 ResNet18, HSJA | trace accuracy: 57.9% -> 99.3% | Improved |
| increase tracer weight alpha from 0.05 to 0.15 (ablation) | separate-and-trace framework · CIFAR10 VGG16, SurFree | trace accuracy: 64.2% -> 87.8% | Improved |
| increase tracer weight alpha from 0 to 0.2 (ablation) | classifier with tracer · mini-ImageNet ResNet50 | classification accuracy: 73.12% -> 68.14% | Hurt |
| increase tracer weight alpha from 0 to 0.2 (ablation) | classifier with tracer · CIFAR10 ResNet18 | classification accuracy: 94.30% -> 93.95% | No clear effect |
| use ResNet instead of VGG as backbone (ablation) | tracing · CIFAR10, Boundary alpha=0.05 | trace accuracy: 92.1% -> 98.1% | Improved |
| geometric attack SurFree instead of gradient-estimation attacks | tracing · CIFAR10 ResNet18, alpha=0.05 | trace accuracy: 98.2% (HSJA) -> 94.5% (SurFree) | Hurt |
| increase number of distributed copies to 10 (ablation) | tracing · CIFAR10 and GTSRB, HSJA alpha=0.15 | trace accuracy | Hurt |
Closest papers by meaning
- Identification of the Adversary from a Single Adversarial Example (ICML 2023 · similarity 0.93)
- Deep Neural Network Fingerprinting by Conferrable Adversarial Examples (ICLR 2021 · similarity 0.87)
- TraceEvader: Making DeepFakes More Untraceable via Evading the Forgery Model Attribution (AAAI 2024 · similarity 0.86)
- Rethinking Adversarial Robustness in the Context of the Right to be Forgotten (ICML 2024 · similarity 0.84)
- Enhancing Adversarial Transferability with Checkpoints of a Single Model's Training (CVPR 2025 · similarity 0.84)
- Attack To Defend: Exploiting Adversarial Attacks for Detecting Poisoned Models (CVPR 2024 · similarity 0.84)
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.
