# Image Classification on ImageNet Subset 64x64 (100 categories)

Source: [Image Classification on ImageNet Subset 64x64 (100 categories)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-subset-64x64-100-categories>)

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

Data updated: 2026-03-13T08:41:43.440Z

Evaluation of Image Classification performance on a specific subset of ImageNet, restricted to 64x64 resolution and the first 100 categories.

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) | ImageNet Subset 64x64 resolution (first 100 categories) |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-subset-64x64-100-categories.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [PGD Robust Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-subset-64x64-100-categories.md?metric=PGD%20Robust%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Accuracy (AA)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-subset-64x64-100-categories.md?metric=Average%20Accuracy%20(AA)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| EB-JDAT-JEM++ | 63.02 | Not recorded | 68fa78c11e209131cea7a4c1 |
| WEAT | 61.9 | Not recorded | 68fa78c11e209131cea7a4c1 |
| LAS-AT | 50.66 | Not recorded | 68fa78c11e209131cea7a4c1 |
| MART | 49.3 | Not recorded | 68fa78c11e209131cea7a4c1 |

### EB-JDAT-JEM++

Accuracy: 63.02. Result ID: 69b3cdb747f34ce9249df0e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78c11e209131cea7a4c1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c11e209131cea7a4c1.md>)

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

### WEAT

Accuracy: 61.9. Result ID: 69b3cdb747f34ce9249df0e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78c11e209131cea7a4c1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74bf1e209131cea63789.md>)

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

### LAS-AT

Accuracy: 50.66. Result ID: 69b3cdb747f34ce9249df0e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78c11e209131cea7a4c1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c001e209131cea31901.md>)

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

### MART

Accuracy: 49.3. Result ID: 69b3cdb747f34ce9249df0e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78c11e209131cea7a4c1.md>)

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

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