# Image Classification on CIFAR-10 (test) (Symmetric & Asymmetric Noise Robustness)

Source: [Image Classification on CIFAR-10 (test) (Symmetric & Asymmetric Noise Robustness)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-and-asymmetric-noise-robustness>)

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

Data updated: 2026-02-26T05:34:27.711Z

Evaluation of Image Classification model robustness on the CIFAR-10 test set using accuracy metrics across various levels of symmetric noise (0.2, 0.5, 0.8, 0.9) and asymmetric noise (0.4).

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 (Symmetric 0.2) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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 (Symmetric 0.2)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-and-asymmetric-noise-robustness.md?metric=Accuracy%20(Symmetric%200.2)>): Higher is better; unit: PERCENT.

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

- [Accuracy (Symmetric 0.8)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-and-asymmetric-noise-robustness.md?metric=Accuracy%20(Symmetric%200.8)>): Higher is better; unit: PERCENT.

- [Accuracy (Symmetric 0.9)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-and-asymmetric-noise-robustness.md?metric=Accuracy%20(Symmetric%200.9)>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Accuracy (Symmetric 0.2) | Result date | Report paper |
| --- | --- | --- | --- |
| CMW-Net-SL+ | 96.6 | Not recorded | 68fa6f331e209131cea44659 |
| AugDesc | 96.2 | Not recorded | 68fa6f331e209131cea44659 |
| C2D | 96.2 | Not recorded | 68fa6f331e209131cea44659 |
| Two-step | 96.2 | Not recorded | 68fa6f331e209131cea44659 |
| CMW-Net-SL | 96.2 | Not recorded | 68fa6f331e209131cea44659 |
| DivideMix | 95.7 | Not recorded | 68fa6f331e209131cea44659 |
| REED | 95.7 | Not recorded | 68fa6f331e209131cea44659 |
| ELR+ | 94.6 | Not recorded | 68fa6f331e209131cea44659 |

### CMW-Net-SL+

Accuracy (Symmetric 0.2): 96.6. Result ID: 69909cc93a6c31484901de95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Implementation](<https://github.com/xjtushujun/CMW-Net>)

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

### AugDesc

Accuracy (Symmetric 0.2): 96.2. Result ID: 69909cc93a6c31484901de91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b21e209131cea2384c.md>)

[Implementation](<https://github.com/KentoNishi/Augmentation-for-LNL>)

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

### C2D

Accuracy (Symmetric 0.2): 96.2. Result ID: 69909cc93a6c31484901de92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bd31e209131cea305cf.md>)

[Implementation](<https://github.com/ContrastToDivide/C2D>)

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

### Two-step

Accuracy (Symmetric 0.2): 96.2. Result ID: 69909cc93a6c31484901de93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

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

### CMW-Net-SL

Accuracy (Symmetric 0.2): 96.2. Result ID: 69909cc93a6c31484901de94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Implementation](<https://github.com/xjtushujun/CMW-Net>)

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

### DivideMix

Accuracy (Symmetric 0.2): 95.7. Result ID: 69909cc93a6c31484901de8e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa671c1e209131cea144cf.md>)

[Implementation](<https://github.com/LiJunnan1992/DivideMix>)

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

### REED

Accuracy (Symmetric 0.2): 95.7. Result ID: 69909cc93a6c31484901de90.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fd91e209131cea479b3.md>)

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

### ELR+

Accuracy (Symmetric 0.2): 94.6. Result ID: 69909cc93a6c31484901de8f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f331e209131cea44659.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d383.md>)

[Implementation](<https://github.com/shengliu66/ELR>)

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.