# Image Classification on ImageNet-C 1.0 (Severity 1)

Source: [Image Classification on ImageNet-C 1.0 (Severity 1)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1>)

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

Data updated: 2026-06-09T09:54:02.063Z

Evaluation of Image Classification performance on the ImageNet-C dataset, version 1.0, corrupted specifically at Severity Level 1. Metrics report performance across 15 corruption types (e.g., Gaussian, Fog, JPEG) and the overall average.

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-C 1.0 (Severity 1) |
| Metric | Accuracy (Gaussian) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Accuracy (Gaussian)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Gaussian)>): Higher is better; unit: PERCENT.

- [Accuracy (Shot)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Shot)>): Higher is better; unit: PERCENT.

- [Accuracy (Impulse)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Impulse)>): Higher is better; unit: PERCENT.

- [Accuracy (Defocus)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Defocus)>): Higher is better; unit: PERCENT.

- [Accuracy (Glass)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Glass)>): Higher is better; unit: PERCENT.

- [Accuracy (Motion)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Motion)>): Higher is better; unit: PERCENT.

- [Accuracy (Zoom)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Zoom)>): Higher is better; unit: PERCENT.

- [Accuracy (Snow)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Snow)>): Higher is better; unit: PERCENT.

- [Accuracy (Frost)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Frost)>): Higher is better; unit: PERCENT.

- [Accuracy (Fog)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Fog)>): Higher is better; unit: PERCENT.

- [Accuracy (Brightness)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Brightness)>): Higher is better; unit: PERCENT.

- [Accuracy (Contrast)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Contrast)>): Higher is better; unit: PERCENT.

- [Accuracy (Elastic)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Elastic)>): Higher is better; unit: PERCENT.

- [Accuracy (Pixelate)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(Pixelate)>): Higher is better; unit: PERCENT.

- [Accuracy (JPEG)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Accuracy%20(JPEG)>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-c-1-0-severity-1.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Gaussian) | Result date | Report paper |
| --- | --- | --- | --- |
| STAR | 60.67 | Not recorded | 6a279e7290966ade919bfb30 |
| STAR (TTA) | 60.24 | Not recorded | 6a279e7290966ade919bfb30 |
| Standard MoE | 59.66 | Not recorded | 6a279e7290966ade919bfb30 |

### STAR

Accuracy (Gaussian): 60.67. Result ID: 6a27e0eefbc373592c25194d.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e7290966ade919bfb30.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a279e7290966ade919bfb30.md>)

Recorded evaluation configuration: {"Backbone":"ViT-S/32","Test-Time Adaptation (TTA)":"false"}. This record does not assert independent reproduction.

### STAR (TTA)

Accuracy (Gaussian): 60.24. Result ID: 6a27e0eefbc373592c25194e.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e7290966ade919bfb30.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a279e7290966ade919bfb30.md>)

Recorded evaluation configuration: {"Backbone":"ViT-S/32","Test-Time Adaptation (TTA)":"true"}. This record does not assert independent reproduction.

### Standard MoE

Accuracy (Gaussian): 59.66. Result ID: 6a27e0eefbc373592c25194c.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e7290966ade919bfb30.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cc11e209131cea35cdf.md>)

Recorded evaluation configuration: {"Backbone":"ViT-S/32","Test-Time Adaptation (TTA)":"false"}. This record does not assert independent reproduction.

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