# Transferability Assessment on CIFAR-10 (test)

Source: [Transferability Assessment on CIFAR-10 (test)](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test>)

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

Data updated: 2026-02-26T06:01:51.546Z

Evaluation of model transferability assessment on the CIFAR-10 test set, utilizing correlation metrics such as Pearson's r and Kendall's tau.

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 | Transferability Assessment |
| Dataset (including any stated split) | CIFAR-10 (test) |
| Metric | Rel@1 |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Rel@1](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Rel%401>): Higher is better; unit: SCALAR.

- [Rel@3](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Rel%403>): Higher is better; unit: SCALAR.

- [Pearson's r](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Pearson%27s%20r>): Higher is better; unit: SCALAR.

- [Kendall's τ](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Kendall%27s%20%CF%84>): Higher is better; unit: SCALAR.

- [Weighted Pearson's r (w)](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Weighted%20Pearson%27s%20r%20(w)>): Higher is better; unit: SCALAR.

- [Weighted Kendall's τ (w)](<https://www.sota2.com/research/sota/transferability-assessment-on-cifar-10-test.md?metric=Weighted%20Kendall%27s%20%CF%84%20(w)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Rel@1 | Result date | Report paper |
| --- | --- | --- | --- |
| LEEP | 1 | Not recorded | 68fa6d171e209131cea379de |
| LogME | 1 | Not recorded | 68fa6d171e209131cea379de |
| NLEEP | 1 | Not recorded | 68fa6d171e209131cea379de |
| SFDA | 1 | Not recorded | 68fa6d171e209131cea379de |

### LEEP

Rel@1: 1. Result ID: 698f8cfb3a6c314849fa060e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682f1e209131cea1a9fc.md>)

Recorded evaluation configuration: {"Model type":"Supervised CNN"}. This record does not assert independent reproduction.

### LogME

Rel@1: 1. Result ID: 698f8cfb3a6c314849fa060f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a381e209131cea26a62.md>)

[Implementation](<https://github.com/thuml/LogME>)

Recorded evaluation configuration: {"Model type":"Supervised CNN"}. This record does not assert independent reproduction.

### NLEEP

Rel@1: 1. Result ID: 698f8cfb3a6c314849fa0610.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d4c1e209131cea38e8c.md>)

[Implementation](<https://github.com/google-research/google-research/tree/master/rank_ckpt>)

Recorded evaluation configuration: {"Model type":"Supervised CNN"}. This record does not assert independent reproduction.

### SFDA

Rel@1: 1. Result ID: 698f8cfb3a6c314849fa0611.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379de.md>)

[Implementation](<https://github.com/TencentARC/SFDA>)

Recorded evaluation configuration: {"Model type":"Supervised CNN"}. This record does not assert independent reproduction.

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