# Image Classification on mini-ImageNet-Blue (80% noise)

Source: [Image Classification on mini-ImageNet-Blue (80% noise)](<https://www.sota2.com/research/sota/image-classification-on-mini-imagenet-blue-80-percent-noise>)

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

Data updated: 2026-02-26T05:28:23.469Z

Evaluation of Image Classification performance on the mini-ImageNet-Blue dataset corrupted with 80% noise.

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) | mini-ImageNet-Blue (80% noise) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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](<https://www.sota2.com/research/sota/image-classification-on-mini-imagenet-blue-80-percent-noise.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Mixup | 51.8 | Not recorded | 68fa6cf91e209131cea37049 |
| Standard | 50.6 | Not recorded | 68fa6cf91e209131cea37049 |
| Standard | 13.4 | Not recorded | 68fa6cf91e209131cea37049 |
| Bootstrap | 13.4 | Not recorded | 68fa6cf91e209131cea37049 |
| NCR | 11.5 | Not recorded | 68fa6cf91e209131cea37049 |
| Label smoothing | 9.1 | Not recorded | 68fa6cf91e209131cea37049 |
| Mixup | 7.8 | Not recorded | 68fa6cf91e209131cea37049 |
| Bootstrap + Mixup | 7.6 | Not recorded | 68fa6cf91e209131cea37049 |
| NCR + Mixup | 7.1 | Not recorded | 68fa6cf91e209131cea37049 |
| Label smoothing + Mixup | 6.8 | Not recorded | 68fa6cf91e209131cea37049 |

### Mixup

Accuracy: 51.8. Result ID: 698f6fdd3a6c314849f88f3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644e1e209131cea03e44.md>)

[Implementation](<https://github.com/facebookresearch/mixup-cifar10>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Oracle: Clean Subset","Data Augmentation":"Mixup"}. This record does not assert independent reproduction.

### Standard

Accuracy: 50.6. Result ID: 698f6fdd3a6c314849f88f3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Oracle: Clean Subset"}. This record does not assert independent reproduction.

### Standard

Accuracy: 13.4. Result ID: 698f6fdd3a6c314849f88f33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline"}. This record does not assert independent reproduction.

### Bootstrap

Accuracy: 13.4. Result ID: 698f6fdd3a6c314849f88f35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61d01e209131ce9f63bb.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline"}. This record does not assert independent reproduction.

### NCR

Accuracy: 11.5. Result ID: 698f6fdd3a6c314849f88f39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Implementation](<https://github.com/google-research/scenic/tree/main/scenic/projects/ncr>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"NCR"}. This record does not assert independent reproduction.

### Label smoothing

Accuracy: 9.1. Result ID: 698f6fdd3a6c314849f88f37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62331e209131ce9f8333.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline"}. This record does not assert independent reproduction.

### Mixup

Accuracy: 7.8. Result ID: 698f6fdd3a6c314849f88f34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644e1e209131cea03e44.md>)

[Implementation](<https://github.com/facebookresearch/mixup-cifar10>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline","Data Augmentation":"Mixup"}. This record does not assert independent reproduction.

### Bootstrap + Mixup

Accuracy: 7.6. Result ID: 698f6fdd3a6c314849f88f36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline","Data Augmentation":"Mixup"}. This record does not assert independent reproduction.

### NCR + Mixup

Accuracy: 7.1. Result ID: 698f6fdd3a6c314849f88f3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

[Implementation](<https://github.com/google-research/scenic/tree/main/scenic/projects/ncr>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"NCR","Data Augmentation":"Mixup"}. This record does not assert independent reproduction.

### Label smoothing + Mixup

Accuracy: 6.8. Result ID: 698f6fdd3a6c314849f88f38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6cf91e209131cea37049.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Protocol":"Baseline","Data Augmentation":"Mixup"}. This record does not assert independent reproduction.

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