# Image Classification on Tiny ImageNet-C noise-free (test)

Source: [Image Classification on Tiny ImageNet-C noise-free (test)](<https://www.sota2.com/research/sota/image-classification-on-tiny-imagenet-c-noise-free-test>)

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

Data updated: 2026-02-26T06:01:52.411Z

Evaluation of Image Classification performance on the clean test set of the Tiny ImageNet-C corruption benchmark suite.

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) | Tiny ImageNet-C noise-free (test) |
| Metric | Relative 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

- [Relative Accuracy](<https://www.sota2.com/research/sota/image-classification-on-tiny-imagenet-c-noise-free-test.md?metric=Relative%20Accuracy>): Higher is better; unit: PERCENT.

- [Mean Corruption Error](<https://www.sota2.com/research/sota/image-classification-on-tiny-imagenet-c-noise-free-test.md?metric=Mean%20Corruption%20Error>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Relative Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| AugMax-DuBIN | 38.66 | Not recorded | 68fa6b881e209131cea2ee6e |
| AugMix | 37.63 | Not recorded | 68fa6b881e209131cea2ee6e |
| ANT | 35.3 | Not recorded | 68fa6b881e209131cea2ee6e |
| Normal | 24.38 | Not recorded | 68fa6b881e209131cea2ee6e |

### AugMax-DuBIN

Relative Accuracy: 38.66. Result ID: 698e9533ffc3901de09eda1a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b881e209131cea2ee6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b881e209131cea2ee6e.md>)

[Implementation](<https://github.com/VITA-Group/AugMax>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### AugMix

Relative Accuracy: 37.63. Result ID: 698e9533ffc3901de09eda18.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b881e209131cea2ee6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67041e209131cea13ccb.md>)

[Implementation](<https://github.com/google-research/augmix>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ANT

Relative Accuracy: 35.3. Result ID: 698e9533ffc3901de09eda19.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b881e209131cea2ee6e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680b1e209131cea19c52.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Normal

Relative Accuracy: 24.38. Result ID: 698e9533ffc3901de09eda17.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b881e209131cea2ee6e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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