# OOD Detection on Far-OOD Average

Source: [OOD Detection on Far-OOD Average](<https://www.sota2.com/research/sota/ood-detection-on-far-ood-average>)

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

Data updated: 2026-04-30T07:24:26.284Z

Evaluation of Out-of-Distribution (OOD) detection performance on the Far-OOD Average configuration, reporting metrics such as AUROC and FPR95.

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 | OOD Detection |
| Dataset (including any stated split) | Far-OOD Average |
| Metric | FPR95 |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 16 of 16 |
| 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

- [FPR95](<https://www.sota2.com/research/sota/ood-detection-on-far-ood-average.md?metric=FPR95>): Lower is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/ood-detection-on-far-ood-average.md?metric=AUROC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | FPR95 | Result date | Report paper |
| --- | --- | --- | --- |
| EPD | 24.77 | Not recorded | 69f2e22e55a3c32af89fc781 |
| SHE | 27.14 | Not recorded | 69f2e22e55a3c32af89fc781 |
| RMDS | 28.75 | Not recorded | 69f2e22e55a3c32af89fc781 |
| ViM | 29.2 | Not recorded | 69f2e22e55a3c32af89fc781 |
| MDS | 29.97 | Not recorded | 69f2e22e55a3c32af89fc781 |
| KNN | 31.93 | Not recorded | 69f2e22e55a3c32af89fc781 |
| GEN | 32.23 | Not recorded | 69f2e22e55a3c32af89fc781 |
| MSP | 51.66 | Not recorded | 69f2e22e55a3c32af89fc781 |
| DICE | 51.76 | Not recorded | 69f2e22e55a3c32af89fc781 |
| TempScale | 53.77 | Not recorded | 69f2e22e55a3c32af89fc781 |
| ReAct | 53.92 | Not recorded | 69f2e22e55a3c32af89fc781 |
| OpenMax | 55.49 | Not recorded | 69f2e22e55a3c32af89fc781 |
| MLS | 79.23 | Not recorded | 69f2e22e55a3c32af89fc781 |
| EBO | 85.34 | Not recorded | 69f2e22e55a3c32af89fc781 |
| GradNorm | 92.64 | Not recorded | 69f2e22e55a3c32af89fc781 |
| ASH | 96.77 | Not recorded | 69f2e22e55a3c32af89fc781 |

### EPD

FPR95: 24.77. Result ID: 69f303a547f34ce924d91149.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### SHE

FPR95: 27.14. Result ID: 69f303a547f34ce924d91146.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### RMDS

FPR95: 28.75. Result ID: 69f303a547f34ce924d91145.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### ViM

FPR95: 29.2. Result ID: 69f303a547f34ce924d91148.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c151e209131cea32091.md>)

[Implementation](<https://github.com/haoqiwang/vim>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### MDS

FPR95: 29.97. Result ID: 69f303a547f34ce924d91140.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

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

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### KNN

FPR95: 31.93. Result ID: 69f303a547f34ce924d9113f.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

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

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### GEN

FPR95: 32.23. Result ID: 69f303a547f34ce924d9113d.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### MSP

FPR95: 51.66. Result ID: 69f303a547f34ce924d91142.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64e41e209131cea0731a.md>)

[Implementation](<https://www.sota2.com/github.com/hendrycks/error-detection>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### DICE

FPR95: 51.76. Result ID: 69f303a547f34ce924d9113b.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### TempScale

FPR95: 53.77. Result ID: 69f303a547f34ce924d91147.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### ReAct

FPR95: 53.92. Result ID: 69f303a547f34ce924d91144.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d8b5.md>)

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### OpenMax

FPR95: 55.49. Result ID: 69f303a547f34ce924d91143.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### MLS

FPR95: 79.23. Result ID: 69f303a547f34ce924d91141.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### EBO

FPR95: 85.34. Result ID: 69f303a547f34ce924d9113c.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### GradNorm

FPR95: 92.64. Result ID: 69f303a547f34ce924d9113e.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6afd1e209131cea2b8e2.md>)

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

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

### ASH

FPR95: 96.77. Result ID: 69f303a547f34ce924d9113a.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22e55a3c32af89fc781.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","ID Accuracy":"81.14%","ID Dataset":"ImageNet-1K"}. This record does not assert independent reproduction.

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