# Image Classification on CIFAR-10 (test) (Symmetric/Asymmetric Accuracy Benchmarking)

Source: [Image Classification on CIFAR-10 (test) (Symmetric/Asymmetric Accuracy Benchmarking)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking>)

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

Data updated: 2026-05-13T08:05:47.293Z

Evaluation of Image Classification model performance on the CIFAR-10 test set, reporting accuracy under specific symmetric and asymmetric perturbation or error thresholds.

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 20%) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 19 of 19 |
| 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 20%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking.md?metric=Accuracy%20(Symmetric%2020%25)>): Higher is better; unit: PERCENT.

- [Accuracy (Symmetric 50%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking.md?metric=Accuracy%20(Symmetric%2050%25)>): Higher is better; unit: PERCENT.

- [Accuracy (Symmetric 80%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking.md?metric=Accuracy%20(Symmetric%2080%25)>): Higher is better; unit: PERCENT.

- [Accuracy (Symmetric 90%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking.md?metric=Accuracy%20(Symmetric%2090%25)>): Higher is better; unit: PERCENT.

- [Accuracy (Asymmetric 40%)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-symmetric-asymmetric-accuracy-benchmarking.md?metric=Accuracy%20(Asymmetric%2040%25)>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Accuracy (Symmetric 20%) | Result date | Report paper |
| --- | --- | --- | --- |
| DivideMix | 96.1 | Not recorded | 69b78e770636e5bf4afbdb99 |
| Mixup | 95.6 | Not recorded | 69b78e770636e5bf4afbdb99 |
| JEC | 95.6 | Not recorded | 69b78e770636e5bf4afbdb99 |
| LossModel | 94 | Not recorded | 69b78e770636e5bf4afbdb99 |
| PENCIL | 92.4 | Not recorded | 69b78e770636e5bf4afbdb99 |
| Coteaching | 89.5 | Not recorded | 69b78e770636e5bf4afbdb99 |
| HamBR | 88.8 | Not recorded | 6a0405b7dbc836b87d827693 |
| RRL | 88.4 | Not recorded | 6a0405b7dbc836b87d827693 |
| FEC | 88.1 | Not recorded | 69b78e770636e5bf4afbdb99 |
| ProMix | 87.4 | Not recorded | 6a0405b7dbc836b87d827693 |
| CE | 86.8 | Not recorded | 69b78e770636e5bf4afbdb99 |
| Forward | 86.8 | Not recorded | 69b78e770636e5bf4afbdb99 |
| L2B | 86.7 | Not recorded | 6a0405b7dbc836b87d827693 |
| LongReMix | 81.4 | Not recorded | 6a0405b7dbc836b87d827693 |
| PSSCL | 80.7 | Not recorded | 6a0405b7dbc836b87d827693 |
| DivideMix | 79.5 | Not recorded | 6a0405b7dbc836b87d827693 |
| GCE | 77.2 | Not recorded | 6a0405b7dbc836b87d827693 |
| Standard CE | 73.4 | Not recorded | 6a0405b7dbc836b87d827693 |
| UNICON | 72.8 | Not recorded | 6a0405b7dbc836b87d827693 |

### DivideMix

Accuracy (Symmetric 20%): 96.1. Result ID: 69b7d75e47f34ce924a01cbb.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.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.

### Mixup

Accuracy (Symmetric 20%): 95.6. Result ID: 69b7d75e47f34ce924a01cbc.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

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

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

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

### JEC

Accuracy (Symmetric 20%): 95.6. Result ID: 69b7d75e47f34ce924a01cbf.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

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

### LossModel

Accuracy (Symmetric 20%): 94. Result ID: 69b7d75e47f34ce924a01cba.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65d71e209131cea0cad4.md>)

[Implementation](<https://git.io/fjsvE>)

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

### PENCIL

Accuracy (Symmetric 20%): 92.4. Result ID: 69b7d75e47f34ce924a01cb9.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65801e209131cea0a8aa.md>)

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

### Coteaching

Accuracy (Symmetric 20%): 89.5. Result ID: 69b7d75e47f34ce924a01cb8.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65bb1e209131cea0bed4.md>)

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

### HamBR

Accuracy (Symmetric 20%): 88.8. Result ID: 6a0430d047f34ce924e9f0eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

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

### RRL

Accuracy (Symmetric 20%): 88.4. Result ID: 6a0430d047f34ce924e9f0e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf21e209131cea31141.md>)

[Implementation](<https://github.com/ShikunLi/Sel-CL>)

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

### FEC

Accuracy (Symmetric 20%): 88.1. Result ID: 69b7d75e47f34ce924a01cbe.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

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

### ProMix

Accuracy (Symmetric 20%): 87.4. Result ID: 6a0430d047f34ce924e9f0e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703e1e209131cea49ea2.md>)

[Implementation](<https://github.com/Justherozen/ProMix>)

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

### CE

Accuracy (Symmetric 20%): 86.8. Result ID: 69b7d75e47f34ce924a01cb7.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

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

### Forward

Accuracy (Symmetric 20%): 86.8. Result ID: 69b7d75e47f34ce924a01cbd.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e770636e5bf4afbdb99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63321e209131ce9fd4c8.md>)

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

### L2B

Accuracy (Symmetric 20%): 86.7. Result ID: 6a0430d047f34ce924e9f0e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

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

### LongReMix

Accuracy (Symmetric 20%): 81.4. Result ID: 6a0430d047f34ce924e9f0e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d5a1e209131cea39623.md>)

[Implementation](<https://github.com/filipe-research/LongReMix>)

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

### PSSCL

Accuracy (Symmetric 20%): 80.7. Result ID: 6a0430d047f34ce924e9f0ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

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

### DivideMix

Accuracy (Symmetric 20%): 79.5. Result ID: 6a0430d047f34ce924e9f0e4.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.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.

### GCE

Accuracy (Symmetric 20%): 77.2. Result ID: 6a0430d047f34ce924e9f0e3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

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

### Standard CE

Accuracy (Symmetric 20%): 73.4. Result ID: 6a0430d047f34ce924e9f0e2.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

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

### UNICON

Accuracy (Symmetric 20%): 72.8. Result ID: 6a0430d047f34ce924e9f0e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405b7dbc836b87d827693.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c691e209131cea33eab.md>)

[Implementation](<https://github.com/nazmul-karim170/UNICON-Noisy-Label>)

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.
