Research map / Adversarial robustness and security / Certified and empirical robustness
Towards More Robust Interpretation via Local Gradient Alignment
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 |
|---|---|---|---|
| add cosine gradient-alignment regularization to ℓ2 regularization (ablation) | LeNet · CIFAR10 | RPS (Grad): 0.881 -> 0.895 | Helped |
| add cosine gradient-alignment regularization to ℓ2 regularization (ablation) | ResNet18 · CIFAR10 | RPS (Grad): 0.908 -> 0.931 | Helped |
| add cosine gradient-alignment regularization to ℓ2 regularization (ablation) | ResNet18 · ImageNet100 | RPS (Grad): 0.913 -> 0.942 | Helped |
| add cosine gradient-alignment regularization to ℓ2 regularization (ablation) | ResNet18 · CIFAR10 | Accuracy: 93.3 -> 93.0 | Hurt |
| add cosine gradient-alignment regularization to ℓ2 regularization (ablation) | ResNet18 · ImageNet100 | Accuracy: 78.0 -> 78.0 | No clear effect |
| train with cosine regularization alone using a large coefficient (ablation) | proposed method · Appendix C experiments | Insertion | Hurt |
Closest papers by meaning
- Robust Models Are More Interpretable Because Attributions Look Normal (ICML 2022 · similarity 0.89)
- Improving Adversarial Robustness of Attribution via Implicit Regularization (ICML 2026 · similarity 0.89)
- Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training (CVPR 2024 · similarity 0.87)
- Pixel-level Certified Explanations via Randomized Smoothing (ICML 2025 · similarity 0.87)
- Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks (ICLR 2025 · similarity 0.87)
- Unlearning-based Neural Interpretations (ICLR 2025 · similarity 0.86)
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.