# Image Classification on CIFAR-10 (test) (Accuracy vs Noise Level)

Source: [Image Classification on CIFAR-10 (test) (Accuracy vs Noise Level)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level>)

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

Data updated: 2026-02-26T05:34:21.699Z

Evaluation of Image Classification performance on the CIFAR-10 test set, specifically measuring accuracy under varying levels of input noise (0.1 to 0.5).

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) | CIFAR-10 (test) |
| Metric | Accuracy (Noise 0.1) |
| 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 (Noise 0.1)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level.md?metric=Accuracy%20(Noise%200.1)>): Higher is better; unit: PERCENT.

- [Accuracy (Noise 0.2)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level.md?metric=Accuracy%20(Noise%200.2)>): Higher is better; unit: PERCENT.

- [Accuracy (Noise 0.3)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level.md?metric=Accuracy%20(Noise%200.3)>): Higher is better; unit: PERCENT.

- [Accuracy (Noise 0.4)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level.md?metric=Accuracy%20(Noise%200.4)>): Higher is better; unit: PERCENT.

- [Accuracy (Noise 0.5)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-vs-noise-level.md?metric=Accuracy%20(Noise%200.5)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Noise 0.1) | Result date | Report paper |
| --- | --- | --- | --- |
| ROLT-DRW | 80.46 | Not recorded | 68fa6a961e209131cea291d1 |
| LDAM-DRW | 79.01 | Not recorded | 68fa6a961e209131cea291d1 |
| ROLT | 77.49 | Not recorded | 68fa6a961e209131cea291d1 |
| LDAM | 75.06 | Not recorded | 68fa6a961e209131cea291d1 |
| BBN | 72.86 | Not recorded | 68fa6a961e209131cea291d1 |
| NCM | 72.37 | Not recorded | 68fa6a961e209131cea291d1 |
| HAR-DRW | 72.12 | Not recorded | 68fa6a961e209131cea291d1 |
| ERM-DRW | 71.99 | Not recorded | 68fa6a961e209131cea291d1 |
| ERM | 69.33 | Not recorded | 68fa6a961e209131cea291d1 |
| cRT | 69.22 | Not recorded | 68fa6a961e209131cea291d1 |

### ROLT-DRW

Accuracy (Noise 0.1): 80.46. Result ID: 699ac3ca5698665ec4520f9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### LDAM-DRW

Accuracy (Noise 0.1): 79.01. Result ID: 699ac3ca5698665ec4520f98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### ROLT

Accuracy (Noise 0.1): 77.49. Result ID: 699ac3ca5698665ec4520f9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### LDAM

Accuracy (Noise 0.1): 75.06. Result ID: 699ac3ca5698665ec4520f97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### BBN

Accuracy (Noise 0.1): 72.86. Result ID: 699ac3ca5698665ec4520f99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67371e209131cea14e59.md>)

[Implementation](<https://github.com/Megvii-Nanjing/BBN>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### NCM

Accuracy (Noise 0.1): 72.37. Result ID: 699ac3ca5698665ec4520f9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa671b1e209131cea144a9.md>)

[Implementation](<https://github.com/facebookresearch/classifier-balancing>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### HAR-DRW

Accuracy (Noise 0.1): 72.12. Result ID: 699ac3ca5698665ec4520f9c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69941e209131cea22d32.md>)

[Implementation](<https://github.com/kaidic/HAR>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### ERM-DRW

Accuracy (Noise 0.1): 71.99. Result ID: 699ac3ca5698665ec4520f96.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### ERM

Accuracy (Noise 0.1): 69.33. Result ID: 699ac3ca5698665ec4520f95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

### cRT

Accuracy (Noise 0.1): 69.22. Result ID: 699ac3ca5698665ec4520f9a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a961e209131cea291d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa671b1e209131cea144a9.md>)

[Implementation](<https://github.com/facebookresearch/classifier-balancing>)

Recorded evaluation configuration: {"imbalance\_ratio":50}. This record does not assert independent reproduction.

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