# OOD Detection on ImageNet-100 1.0 (test)

Source: [OOD Detection on ImageNet-100 1.0 (test)](<https://www.sota2.com/research/sota/ood-detection-on-imagenet-100-1-0-test>)

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

Data updated: 2026-02-26T05:30:09.382Z

Evaluation of Out-of-Distribution (OOD) detection performance using the ImageNet-100 dataset (version 1.0, test split) as the in-distribution set. Metrics reported include Accuracy (ACC), False Positive Rate at 95% True Positive Rate (FPR95), and Area Under the ROC Curve (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 | OOD Detection |
| Dataset (including any stated split) | ImageNet-100 1.0 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/ood-detection-on-imagenet-100-1-0-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [FPR95](<https://www.sota2.com/research/sota/ood-detection-on-imagenet-100-1-0-test.md?metric=FPR95>): Lower is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/ood-detection-on-imagenet-100-1-0-test.md?metric=AUROC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MSP | 94.77 | Not recorded | 68fa75ae1e209131cea693fe |
| Energy | 94.77 | Not recorded | 68fa75ae1e209131cea693fe |
| VOS | 94.75 | Not recorded | 68fa75ae1e209131cea693fe |
| NPOS | 94.34 | Not recorded | 68fa75ae1e209131cea693fe |
| CATEX | 94.11 | Not recorded | 68fa75ae1e209131cea693fe |

### MSP

Accuracy: 94.77. Result ID: 6993f5483a6c31484919cc41.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a701e209131cea280e5.md>)

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

Recorded evaluation configuration: {"Backbone":"CLIP-B/16","Number of runs":3}. This record does not assert independent reproduction.

### Energy

Accuracy: 94.77. Result ID: 6993f5483a6c31484919cc42.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

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

Recorded evaluation configuration: {"Backbone":"CLIP-B/16","Number of runs":3}. This record does not assert independent reproduction.

### VOS

Accuracy: 94.75. Result ID: 6993f5483a6c31484919cc43.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c831e209131cea346d4.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/16","Number of runs":3}. This record does not assert independent reproduction.

### NPOS

Accuracy: 94.34. Result ID: 6993f5483a6c31484919cc44.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41aaa.md>)

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

Recorded evaluation configuration: {"Backbone":"CLIP-B/16","Number of runs":3}. This record does not assert independent reproduction.

### CATEX

Accuracy: 94.11. Result ID: 6993f5483a6c31484919cc45.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75ae1e209131cea693fe.md>)

[Implementation](<https://github.com/alibaba/catex>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/16","Number of runs":3}. This record does not assert independent reproduction.

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