# Adversarial Robustness on CIFAR-10 (L-infinity, epsilon=8/255)

Source: [Adversarial Robustness on CIFAR-10 (L-infinity, epsilon=8/255)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:02:21.951Z

Evaluation of model adversarial robustness on the CIFAR-10 dataset specifically constrained by an L-infinity perturbation budget of epsilon=8/255, measured using metrics related to AA and A3 attack methodologies.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Adversarial Robustness |
| Dataset (including any stated split) | CIFAR-10 L-infinity, epsilon=8/255 |
| Metric | Robust Acc (AA) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Robust Acc (AA)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Robust%20Acc%20(AA)>): Higher is better; unit: PERCENT.

- [Robust Acc (A3)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Robust%20Acc%20(A3)>): Higher is better; unit: PERCENT.

- [Robust Acc Delta](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Robust%20Acc%20Delta>): Higher is better; unit: SCALAR.

- [Runtime (AA)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Runtime%20(AA)>): Lower is better; unit: TIME.

- [Runtime (A3)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Runtime%20(A3)>): Lower is better; unit: TIME.

- [Speed-up](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Speed-up>): Higher is better; unit: SCALAR.

- [Search Steps (A3)](<https://www.sota2.com/research/sota/adversarial-robustness-on-cifar-10-l-infinity-epsilon-8-255.md?metric=Search%20Steps%20(A3)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Robust Acc (AA) | Result date | Report paper |
| --- | --- | --- | --- |
| Gowal et al. (2021) | 62.8 | Not recorded | 68fa6b161e209131cea2c0c7 |
| Wu et al. (2020) RTS | 60.04 | Not recorded | 68fa6b161e209131cea2c0c7 |
| Zhang et al. (2021) | 59.64 | Not recorded | 68fa6b161e209131cea2c0c7 |
| Carmon et al. (2019) | 59.53 | Not recorded | 68fa6b161e209131cea2c0c7 |
| Sehwag et al. (2020) | 57.14 | Not recorded | 68fa6b161e209131cea2c0c7 |

### Gowal et al. (2021)

Robust Acc (AA): 62.8. Result ID: 698e5513ffc3901de09cb203.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69981e209131cea22f05.md>)

Recorded evaluation configuration: {"author\_provided":true}. This record does not assert independent reproduction.

### Wu et al. (2020) RTS

Robust Acc (AA): 60.04. Result ID: 698e5513ffc3901de09cb204.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa687e1e209131cea1c519.md>)

Recorded evaluation configuration: {"author\_provided":true}. This record does not assert independent reproduction.

### Zhang et al. (2021)

Robust Acc (AA): 59.64. Result ID: 698e5513ffc3901de09cb205.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25c57.md>)

Recorded evaluation configuration: {"author\_provided":true,"epsilon":0.031}. This record does not assert independent reproduction.

### Carmon et al. (2019)

Robust Acc (AA): 59.53. Result ID: 698e5513ffc3901de09cb206.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b9d1e209131cea2f566.md>)

Recorded evaluation configuration: {"author\_provided":true}. This record does not assert independent reproduction.

### Sehwag et al. (2020)

Robust Acc (AA): 57.14. Result ID: 698e5513ffc3901de09cb207.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0c7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68bd1e209131cea1df75.md>)

Recorded evaluation configuration: {"author\_provided":true}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.