# Image Classification on ImageNet (test) (Corruption and Adversarial Robustness)

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

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

Data updated: 2026-02-26T05:26:50.451Z

Evaluation of Image Classification performance on the test split of ImageNet, reporting robustness metrics against common visual corruptions (e.g., Hue, Saturation, Brightness) and specific adversarial perturbations (e.g., l\_inf attacks).

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 (test) |
| Metric | Acc (Clean) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 0 of 0 |
| 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

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

- [Acc (Hue Corruption)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Hue%20Corruption)>): Higher is better; unit: PERCENT.

- [Acc (Saturation Corruption)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Saturation%20Corruption)>): Higher is better; unit: PERCENT.

- [Acc (Rotation Corruption)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Rotation%20Corruption)>): Higher is better; unit: PERCENT.

- [Acc (Brightness Corruption)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Brightness%20Corruption)>): Higher is better; unit: PERCENT.

- [Acc (Contrast Corruption)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Contrast%20Corruption)>): Higher is better; unit: PERCENT.

- [Acc (l\_inf 20-step Attack)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(l_inf%2020-step%20Attack)>): Higher is better; unit: PERCENT.

- [Acc (Auto-l\_inf Attack)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-corruption-and-adversarial-robustness.md?metric=Acc%20(Auto-l_inf%20Attack)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Acc (Clean) | Result date | Report paper |
| --- | --- | --- | --- |

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