# Image Classification on CIFAR-10 W (OOD Evaluation)

Source: [Image Classification on CIFAR-10 W (OOD Evaluation)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-w-ood-evaluation>)

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

Data updated: 2026-04-01T07:32:14.422Z

Evaluation of Image Classification performance on the CIFAR-10 W dataset, focusing specifically on Out-of-Distribution (OOD) metrics.

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 | Image Classification |
| Dataset (including any stated split) | CIFAR-10 W |
| Metric | OOD Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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

- [OOD Accuracy](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-w-ood-evaluation.md?metric=OOD%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | OOD Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| HIRPG + CONAN | 77.64 | Not recorded | 68fa75521e209131cea67246 |
| CHB + CONAN | 75.96 | Not recorded | 68fa75521e209131cea67246 |
| LE + CONAN | 75.13 | Not recorded | 68fa75521e209131cea67246 |
| SC + CONAN | 75.04 | Not recorded | 68fa75521e209131cea67246 |
| HIRPG | 74.47 | Not recorded | 68fa75521e209131cea67246 |
| LE | 73.51 | Not recorded | 68fa75521e209131cea67246 |
| SC | 71.3 | Not recorded | 68fa75521e209131cea67246 |
| CHB | 70.61 | Not recorded | 68fa75521e209131cea67246 |
| CCG + CONAN | 63.14 | Not recorded | 68fa75521e209131cea67246 |
| Manually Annotated | 59.12 | Not recorded | 68fa75521e209131cea67246 |
| CCG | 58.25 | Not recorded | 68fa75521e209131cea67246 |
| CLIP-ZS | 57.14 | Not recorded | 68fa75521e209131cea67246 |
| Glide-syn | 55.93 | Not recorded | 68fa75521e209131cea67246 |
| SuS-X-SD | 53.08 | Not recorded | 68fa75521e209131cea67246 |
| SD-Clf | 52.48 | Not recorded | 68fa75521e209131cea67246 |
| VisDesc | 51.83 | Not recorded | 68fa75521e209131cea67246 |
| CALIP | 51.62 | Not recorded | 68fa75521e209131cea67246 |
| CuPL | 50.5 | Not recorded | 68fa75521e209131cea67246 |

### HIRPG + CONAN

OOD Accuracy: 77.64. Result ID: 69ccc93947f34ce924b61d88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### CHB + CONAN

OOD Accuracy: 75.96. Result ID: 69ccc93947f34ce924b61d82.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### LE + CONAN

OOD Accuracy: 75.13. Result ID: 69ccc93947f34ce924b61d80.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### SC + CONAN

OOD Accuracy: 75.04. Result ID: 69ccc93947f34ce924b61d84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### HIRPG

OOD Accuracy: 74.47. Result ID: 69ccc93947f34ce924b61d87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### LE

OOD Accuracy: 73.51. Result ID: 69ccc93947f34ce924b61d7f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### SC

OOD Accuracy: 71.3. Result ID: 69ccc93947f34ce924b61d83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### CHB

OOD Accuracy: 70.61. Result ID: 69ccc93947f34ce924b61d81.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### CCG + CONAN

OOD Accuracy: 63.14. Result ID: 69ccc93947f34ce924b61d86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### Manually Annotated

OOD Accuracy: 59.12. Result ID: 69ccc93947f34ce924b61d89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### CCG

OOD Accuracy: 58.25. Result ID: 69ccc93947f34ce924b61d85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### CLIP-ZS

OOD Accuracy: 57.14. Result ID: 69ccc93947f34ce924b61d78.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

### Glide-syn

OOD Accuracy: 55.93. Result ID: 69ccc93947f34ce924b61d7e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8b1e209131cea40739.md>)

Recorded evaluation configuration: {"Type":"Training-dependent"}. This record does not assert independent reproduction.

### SuS-X-SD

OOD Accuracy: 53.08. Result ID: 69ccc93947f34ce924b61d79.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa705a1e209131cea4a85f.md>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

### SD-Clf

OOD Accuracy: 52.48. Result ID: 69ccc93947f34ce924b61d7d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa708c1e209131cea4baff.md>)

[Implementation](<https://www.sota2.com/diffusion-classifier.github.io/>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

### VisDesc

OOD Accuracy: 51.83. Result ID: 69ccc93947f34ce924b61d7a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e221e209131cea3e10c.md>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

### CALIP

OOD Accuracy: 51.62. Result ID: 69ccc93947f34ce924b61d7c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e421e209131cea3ec42.md>)

[Implementation](<https://github.com/ZiyuGuo99/CALIP>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

### CuPL

OOD Accuracy: 50.5. Result ID: 69ccc93947f34ce924b61d7b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75521e209131cea67246.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71551e209131cea508eb.md>)

Recorded evaluation configuration: {"Type":"Training-free"}. This record does not assert independent reproduction.

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