Research map / Adversarial robustness and security / Certified and empirical robustness
Detecting Adversarial Examples from Sensitivity Inconsistency of Spatial-Transform Domain
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the WAWT dual classifier and sensitivity-inconsistency detector instead of MD | ResNet adversarial-example detection · SVHN, matching-source DeepFool | AUC (%): 95.32 -> 98.15 | Helped |
| Use the WAWT dual classifier and sensitivity-inconsistency detector instead of MD | ResNet adversarial-example detection · SVHN, matching-source BIM | AUC (%): 89.69 -> 99.47 | Helped |
| Use the WAWT dual classifier and sensitivity-inconsistency detector instead of MD | ResNet adversarial-example detection · CIFAR10, matching-source BIM | AUC (%): 87.01 -> 99.68 | Helped |
| Train SID on one attack source and test it on other attack types | ResNet adversarial-example detection · SVHN and CIFAR10, cross-attack generalization | AUC (%) | Helped |
| Use SID for FGSM examples at increasing perturbation magnitudes | ResNet-SVHN adversarial-example detection · FGSM, η > 0.41 | AUC | Mixed |
| Evaluate SID on BIM and DeepFool examples across perturbation magnitudes | ResNet-SVHN adversarial-example detection · BIM and DeepFool, all tested η | AUC | Helped |
| Attack both the target classifier and SID with a white-box objective | SID trained with BIM-L · ResNet-CIFAR10 | AUC: 99.81 -> 98.31 | Hurt |
| Attack both the target classifier and SID with a white-box objective | SID trained with BIM-L · ResNet-SVHN | AUC: 98.51 -> 98.13 | Hurt |
Closest papers by meaning
- Decision Boundary Analysis of Adversarial Examples (ICLR 2018 · similarity 0.87)
- Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations (ICML 2020 · similarity 0.86)
- Universal Adversarial Perturbations (CVPR 2017 · similarity 0.86)
- Doppelgangers and Adversarial Vulnerability (CVPR 2025 · similarity 0.85)
- One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancy (ICML 2025 · similarity 0.85)
- Be Your Own Neighborhood: Detecting Adversarial Examples by the Neighborhood Relations Built on Self-Supervised Learning (ICML 2024 · similarity 0.85)
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.