# Out-of-Distribution Detection on CIFAR-10 vs CIFAR-100 Near-OOD (test)

Source: [Out-of-Distribution Detection on CIFAR-10 vs CIFAR-100 Near-OOD (test)](<https://www.sota2.com/research/sota/out-of-distribution-detection-on-cifar-10-vs-cifar-100-near-ood-test>)

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

Data updated: 2026-02-26T05:25:53.725Z

Evaluation of Out-of-Distribution Detection performance using the Near-OOD setup comparing CIFAR-10 and CIFAR-100 on the test split, typically measured using AUROC.

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 | Out-of-Distribution Detection |
| Dataset (including any stated split) | CIFAR-10 vs CIFAR-100 Near-OOD (test) |
| Metric | AUROC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [AUROC](<https://www.sota2.com/research/sota/out-of-distribution-detection-on-cifar-10-vs-cifar-100-near-ood-test.md?metric=AUROC>): Higher is better; unit: PERCENT.

- [FPR@95%TPR](<https://www.sota2.com/research/sota/out-of-distribution-detection-on-cifar-10-vs-cifar-100-near-ood-test.md?metric=FPR%4095%25TPR>): Lower is better; unit: PERCENT.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| shifting transformation learning | 94.07 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| Outlier Exposure | 93.41 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| CSI-ens | 92.06 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| Geometric | 91.91 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| SSD | 90.4 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| KNN | 89.55 | Not recorded | 68fa70ad1e209131cea4c735 |
| Mahalanobis | 88.51 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| NNGuide | 86.64 | Not recorded | 68fa70ad1e209131cea4c735 |
| Energy | 86.15 | Not recorded | 68fa70ad1e209131cea4c735 |
| ODIN | 85.8 | Not recorded | 68fa6afc1e209131cea2b8a7 |
| Gram Matrices | 79.01 | Not recorded | 68fa6afc1e209131cea2b8a7 |

### shifting transformation learning

AUROC: 94.07. Result ID: 698e49afffc3901de09c55ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"No","Generalizable":true,"Zero Shot":true}. This record does not assert independent reproduction.

### Outlier Exposure

AUROC: 93.41. Result ID: 698e49afffc3901de09c55e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654c1e209131cea096af.md>)

[Implementation](<https://github.com/hendrycks/outlier-exposure>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"Low","Generalizable":true,"Zero Shot":false}. This record does not assert independent reproduction.

### CSI-ens

AUROC: 92.06. Result ID: 698e49afffc3901de09c55e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68881e209131cea1cb66.md>)

[Implementation](<https://github.com/alinlab/CSI>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"Low","Generalizable":true,"Zero Shot":true}. This record does not assert independent reproduction.

### Geometric

AUROC: 91.91. Result ID: 698e49afffc3901de09c55e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65161e209131cea082d3.md>)

[Implementation](<https://github.com/izikgo/AnomalyDetectionTransformations>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"Low","Generalizable":true,"Zero Shot":true}. This record does not assert independent reproduction.

### SSD

AUROC: 90.4. Result ID: 698e49afffc3901de09c55ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa698f1e209131cea22a16.md>)

[Implementation](<https://github.com/inspire-group/SSD>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"No","Generalizable":true,"Zero Shot":true}. This record does not assert independent reproduction.

### KNN

AUROC: 89.55. Result ID: 6991512c3a6c31484907452c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70ad1e209131cea4c735.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e2f1e209131cea3e69b.md>)

[Implementation](<https://github.com/deeplearning-wisc/knn-ood>)

Recorded evaluation configuration: {"k":10}. This record does not assert independent reproduction.

### Mahalanobis

AUROC: 88.51. Result ID: 698e49afffc3901de09c55e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64f91e209131cea077ac.md>)

[Implementation](<https://github.com/pokaxpoka/deep_Mahalanobis_detector>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"High","Generalizable":false,"Zero Shot":false}. This record does not assert independent reproduction.

### NNGuide

AUROC: 86.64. Result ID: 6991512c3a6c31484907452d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70ad1e209131cea4c735.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ad1e209131cea4c735.md>)

[Implementation](<https://github.com/roomo7time/nnguide>)

Recorded evaluation configuration: {"alpha":"10%","k":10}. This record does not assert independent reproduction.

### Energy

AUROC: 86.15. Result ID: 6991512c3a6c31484907452b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70ad1e209131cea4c735.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69cc1e209131cea24242.md>)

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

### ODIN

AUROC: 85.8. Result ID: 698e49afffc3901de09c55e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64141e209131cea029d7.md>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"High","Generalizable":false,"Zero Shot":false}. This record does not assert independent reproduction.

### Gram Matrices

AUROC: 79.01. Result ID: 698e49afffc3901de09c55eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6afc1e209131cea2b8a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66e31e209131cea12d95.md>)

[Implementation](<https://github.com/VectorInstitute/gram-ood-detection>)

Recorded evaluation configuration: {"Hyp.-Para. Dependency":"No","Generalizable":true,"Zero Shot":true}. This record does not assert independent reproduction.

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