Papers with Code · CC BY-SA 4.0

Adversarial Robustness

33 평가 기록 7 datasets

Adversarial Robustness evaluates the vulnerabilities of machine learning models under various types of adversarial attacks.

논문

Are Transformers More Robust Than CNNs?

2111.05464 2021-11-10 이 페이지의 평가 16건

Accuracy: 77.4

Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models

2111.02840 2021-11-04 이 페이지의 평가 10건

Accuracy: 0.6086

Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing

2301.12554 2023-01-29 이 페이지의 평가 2건

Clean Accuracy: 85.21AutoAttacked Accuracy: 38.72

Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria

2310.03358 2023-10-05 이 페이지의 평가 1건

Clean Accuracy: 60.10AutoAttacked Accuracy: 35.05

Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness

2112.02671 2021-12-05 이 페이지의 평가 1건

Attack: AutoAttack: 82.6Accuracy: 92.26Robust Accuracy: 84.3