# OOD Detection on ImageNet-1k (ID) vs ImageNet-O (OOD)

Source: [OOD Detection on ImageNet-1k (ID) vs ImageNet-O (OOD)](<https://www.sota2.com/research/sota/ood-detection-on-imagenet-1k-id-vs-imagenet-o-ood>)

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

Data updated: 2026-05-26T10:30:00.071Z

Evaluation of Out-of-Distribution (OOD) detection performance where ImageNet-1k serves as the in-distribution dataset and ImageNet-O serves as the out-of-distribution dataset. Key metrics reported are 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) | ImageNet-1k (ID) ImageNet-O (OOD) |
| Metric | AUROC |
| Unit | PERCENT |
| 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

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

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

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| ∆Energy | 90.49 | Not recorded | 68fa7c1e1e209131cea8c803 |
| CSP | 88.08 | Not recorded | 68fa7c1e1e209131cea8c803 |
| NegLabel | 85.78 | Not recorded | 68fa7c1e1e209131cea8c803 |
| MCM | 78.59 | Not recorded | 68fa7c1e1e209131cea8c803 |

### ∆Energy

AUROC: 90.49. Result ID: 6a15760afbc373592c0d871b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c803.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c803.md>)

Recorded evaluation configuration: {"mode":"zero-shot"}. This record does not assert independent reproduction.

### CSP

AUROC: 88.08. Result ID: 6a15760afbc373592c0d871a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c803.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75351e209131cea666f2.md>)

Recorded evaluation configuration: {"mode":"zero-shot"}. This record does not assert independent reproduction.

### NegLabel

AUROC: 85.78. Result ID: 6a15760afbc373592c0d8719.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c803.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72b01e209131cea57e55.md>)

Recorded evaluation configuration: {"mode":"zero-shot"}. This record does not assert independent reproduction.

### MCM

AUROC: 78.59. Result ID: 6a15760afbc373592c0d8718.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c803.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e0a1e209131cea3d9c6.md>)

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

Recorded evaluation configuration: {"mode":"zero-shot"}. This record does not assert independent reproduction.

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