# Image Classification on ImageNet (Adversarial Robustness Metrics)

Source: [Image Classification on ImageNet (Adversarial Robustness Metrics)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-adversarial-robustness-metrics>)

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

Data updated: 2026-02-27T11:46:09.188Z

Evaluation of Image Classification performance on the standard ImageNet dataset, specifically focusing on robustness against adversarial attacks, measured by Clean accuracy, AutoAttack results, and PGD attack results.

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 | Image Classification |
| Dataset (including any stated split) | ImageNet |
| Metric | Clean Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Clean Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-adversarial-robustness-metrics.md?metric=Clean%20Accuracy>): Higher is better; unit: PERCENT.

- [Robust Accuracy (AutoAttack)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-adversarial-robustness-metrics.md?metric=Robust%20Accuracy%20(AutoAttack)>): Higher is better; unit: PERCENT.

- [Robust Accuracy (PGD)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-adversarial-robustness-metrics.md?metric=Robust%20Accuracy%20(PGD)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Clean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FARE | 85.01 | Not recorded | 6940e7a8456146936284f897 |
| UCAT | 70.48 | Not recorded | 6940e7a8456146936284f897 |
| TeCoA | 63.87 | Not recorded | 6940e7a8456146936284f897 |
| CLIP | 59.15 | Not recorded | 6940e7a8456146936284f897 |

### FARE

Clean Accuracy: 85.01. Result ID: 69a183f135dfcabb00e788c7.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d24a.md>)

Recorded evaluation configuration: {"Training Data":"ImageNet-1k","Adversarial Training":"2-step PGD","Epsilon":"1/255"}. This record does not assert independent reproduction.

### UCAT

Clean Accuracy: 70.48. Result ID: 69a183f135dfcabb00e788c8.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

Recorded evaluation configuration: {"Training Data":"ImageNet-1k","Adversarial Training":"2-step PGD","Epsilon":"1/255"}. This record does not assert independent reproduction.

### TeCoA

Clean Accuracy: 63.87. Result ID: 69a183f135dfcabb00e788c6.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f271e209131cea44177.md>)

Recorded evaluation configuration: {"Training Data":"ImageNet-1k","Adversarial Training":"2-step PGD","Epsilon":"1/255"}. This record does not assert independent reproduction.

### CLIP

Clean Accuracy: 59.15. Result ID: 69a183f135dfcabb00e788c5.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Training Data":"ImageNet-1k","Adversarial Training":"2-step PGD","Epsilon":"1/255"}. This record does not assert independent reproduction.

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