# Image Classification on ImageNet (Corruption Robustness)

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

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

Data updated: 2026-02-26T05:39:22.419Z

Evaluation of Image Classification performance on the standard ImageNet dataset, specifically measuring robustness against various input corruptions and perturbations, including color shifts (Hue, Saturation, Brightness, Contrast) and adversarial metrics (l-infinity norms).

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 Error Rate |
| Unit | PERCENT |
| Direction | Lower 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

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

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

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

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

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

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

- [L-inf (20-step) Error](<https://www.sota2.com/research/sota/image-classification-on-imagenet-corruption-robustness.md?metric=L-inf%20(20-step)%20Error>): Lower is better; unit: PERCENT.

- [Auto L-inf Error](<https://www.sota2.com/research/sota/image-classification-on-imagenet-corruption-robustness.md?metric=Auto%20L-inf%20Error>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Clean Error Rate | Result date | Report paper |
| --- | --- | --- | --- |

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