# Image Classification on Colored MNIST 5.0% bias-conflict ratio

Source: [Image Classification on Colored MNIST 5.0% bias-conflict ratio](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio>)

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

Data updated: 2026-05-28T10:02:18.824Z

Evaluation of Image Classification performance on the Colored MNIST dataset configured with a 5.0% bias-conflict ratio.

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) | Colored MNIST 5.0% bias-conflict ratio |
| Metric | BC Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [BC Score](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio.md?metric=BC%20Score>): Higher is better; unit: PERCENT.

- [BA Score](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio.md?metric=BA%20Score>): Higher is better; unit: PERCENT.

- [Average Performance](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio.md?metric=Average%20Performance>): Higher is better; unit: PERCENT.

- [Best BC Score](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio.md?metric=Best%20BC%20Score>): Higher is better; unit: PERCENT.

- [BA at Best BC](<https://www.sota2.com/research/sota/image-classification-on-colored-mnist-5-0-percent-bias-conflict-ratio.md?metric=BA%20at%20Best%20BC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BC Score | Result date | Report paper |
| --- | --- | --- | --- |
| BiasEdit | 91.72 | Not recorded | 6a17cc4bdbd7150506f81547 |
| SelecMix | 89.46 | Not recorded | 6a17cc4bdbd7150506f81547 |
| LfF | 86.89 | Not recorded | 6a17cc4bdbd7150506f81547 |
| DFA | 85.71 | Not recorded | 6a17cc4bdbd7150506f81547 |
| DeNetDM | 80.72 | Not recorded | 6a17cc4bdbd7150506f81547 |
| BiasAdv | 78.31 | Not recorded | 6a17cc4bdbd7150506f81547 |
| Vanilla | 76.79 | Not recorded | 6a17cc4bdbd7150506f81547 |
| LfF+BE | 75.18 | Not recorded | 6a17cc4bdbd7150506f81547 |
| SoftCon | 68.66 | Not recorded | 6a17cc4bdbd7150506f81547 |

### BiasEdit

BC Score: 91.72. Result ID: 6a181292fbc373592c12579b.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

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

### SelecMix

BC Score: 89.46. Result ID: 6a181292fbc373592c125798.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddf1e209131cea3c994.md>)

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

### LfF

BC Score: 86.89. Result ID: 6a181292fbc373592c125794.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

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

### DFA

BC Score: 85.71. Result ID: 6a181292fbc373592c125797.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b191e209131cea2c2bc.md>)

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

### DeNetDM

BC Score: 80.72. Result ID: 6a181292fbc373592c12579a.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758c1e209131cea68895.md>)

[Implementation](<https://github.com/kadarsh22/DeNetDM>)

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

### BiasAdv

BC Score: 78.31. Result ID: 6a181292fbc373592c125799.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

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

### Vanilla

BC Score: 76.79. Result ID: 6a181292fbc373592c125793.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

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

### LfF+BE

BC Score: 75.18. Result ID: 6a181292fbc373592c125795.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e201e209131cea3df92.md>)

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

### SoftCon

BC Score: 68.66. Result ID: 6a181292fbc373592c125796.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc4bdbd7150506f81547.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.
