# Unsupervised Model Selection on CIFAR-10-C

Source: [Unsupervised Model Selection on CIFAR-10-C](<https://www.sota2.com/research/sota/unsupervised-model-selection-on-cifar-10-c>)

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

Data updated: 2026-05-12T10:01:17.457Z

Evaluation of unsupervised model selection performance on the CIFAR-10-C corrupted dataset, typically measured using Spearman rho and selection accuracy.

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 | Unsupervised Model Selection |
| Dataset (including any stated split) | CIFAR-10-C |
| Metric | Spearman Rho |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Spearman Rho](<https://www.sota2.com/research/sota/unsupervised-model-selection-on-cifar-10-c.md?metric=Spearman%20Rho>): Higher is better; unit: SCALAR.

- [Selection Accuracy](<https://www.sota2.com/research/sota/unsupervised-model-selection-on-cifar-10-c.md?metric=Selection%20Accuracy>): Higher is better; unit: PERCENT.

- [Oracle Accuracy](<https://www.sota2.com/research/sota/unsupervised-model-selection-on-cifar-10-c.md?metric=Oracle%20Accuracy>): Higher is better; unit: PERCENT.

- [Performance Gap](<https://www.sota2.com/research/sota/unsupervised-model-selection-on-cifar-10-c.md?metric=Performance%20Gap>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Spearman Rho | Result date | Report paper |
| --- | --- | --- | --- |
| Ricci-only | 0.265 | Not recorded | 6a02b407dbc836b87d82460f |
| Fixed GeoScore | -0.415 | Not recorded | 6a02b407dbc836b87d82460f |
| Topology-aware fixed | -0.622 | Not recorded | 6a02b407dbc836b87d82460f |
| Torsion-only | -0.629 | Not recorded | 6a02b407dbc836b87d82460f |
| TopoGeoScore (SSL) | -0.641 | Not recorded | 6a02b407dbc836b87d82460f |

### Ricci-only

Spearman Rho: 0.265. Result ID: 6a02fa6c47f34ce924e6413a.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.md>)

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

### Fixed GeoScore

Spearman Rho: -0.415. Result ID: 6a02fa6c47f34ce924e6413b.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.md>)

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

### Topology-aware fixed

Spearman Rho: -0.622. Result ID: 6a02fa6c47f34ce924e6413c.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.md>)

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

### Torsion-only

Spearman Rho: -0.629. Result ID: 6a02fa6c47f34ce924e64139.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.md>)

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

### TopoGeoScore (SSL)

Spearman Rho: -0.641. Result ID: 6a02fa6c47f34ce924e6413d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b407dbc836b87d82460f.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.
