Research map / Adversarial robustness and security / Black-box and transfer attacks
RobustBench: a standardized adversarial robustness benchmark
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 |
|---|---|---|---|
| Square Attack with 30 random restarts instead of 1 (adaptive attack) | flagged defense · CIFAR-10 Linf | robust accuracy: 18.50% -> 7.40% | Helped |
| standardized AutoAttack evaluation vs original paper evaluations | published defenses · RobustBench leaderboards | robust accuracy estimate | Helped |
| use extra training data | adversarially trained models · CIFAR-10 Linf | robustness-accuracy trade-off | Helped |
| Linf adversarial training | CIFAR-10 classifiers · CIFAR-10 | ECE (calibration) | Hurt |
| Linf adversarial training | CIFAR-10 classifiers · CIFAR-10-C / CIFAR-10.1 / CINIC-10 | accuracy under distribution shift | Helped |
| robust training | CIFAR-10 classifiers · CIFAR-10 | membership inference privacy leakage | Mixed |
| random seed of AutoAttack | Model Zoo models · CIFAR-10/100, ImageNet | robust accuracy | No clear effect |
Closest papers by meaning
- OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift (ICML 2024 · similarity 0.92)
- TabularBench: Benchmarking Adversarial Robustness for Tabular Deep Learning in Real-world Use-cases (NeurIPS 2024 · similarity 0.91)
- Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks (CVPR 2025 · similarity 0.90)
- ROME is Forged in Adversity: Robust Distilled Datasets via Information Bottleneck (ICML 2025 · similarity 0.90)
- GREAT Score: Global Robustness Evaluation of Adversarial Perturbation using Generative Models (NeurIPS 2024 · similarity 0.90)
- Efficient Robustness Evaluation via Constraint Relaxation (AAAI 2025 · similarity 0.90)
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.