# Image Classification on CIFAR-10 resized to 224 × 224 (test)

Source: [Image Classification on CIFAR-10 resized to 224 × 224 (test)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-resized-to-224-224-test>)

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

Data updated: 2026-02-26T05:28:38.259Z

Evaluation of Image Classification performance on the CIFAR-10 test set after resizing the input images to 224x224 resolution.

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 resized to 224 × 224 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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-cifar-10-resized-to-224-224-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| BEiT large | 97.4 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| BEiT large | 97.2 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| BEiT large | 97.1 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| BEiT large | 96.7 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| De et al. | 96.6 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| CrossViT base | 96.5 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| CrossViT base | 96.2 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| De et al. | 96.1 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| CrossViT base | 96.1 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| CrossViT base | 95.5 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| De et al. | 95.4 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| Yu et al. | 94.8 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| De et al. | 94.8 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| Yu et al. | 94.3 | Not recorded | 68fa6e1b1e209131cea3dce4 |
| Tramer et al. | 92.7 | Not recorded | 68fa6e1b1e209131cea3dce4 |

### BEiT large

Accuracy: 97.4. Result ID: 699002963a6c314849fd7285.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"8","pretrained\_dataset":"ImageNet21k","parameters":"303M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### BEiT large

Accuracy: 97.2. Result ID: 699002963a6c314849fd7284.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"4","pretrained\_dataset":"ImageNet21k","parameters":"303M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### BEiT large

Accuracy: 97.1. Result ID: 699002963a6c314849fd7283.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"2","pretrained\_dataset":"ImageNet21k","parameters":"303M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### BEiT large

Accuracy: 96.7. Result ID: 699002963a6c314849fd7282.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"1","pretrained\_dataset":"ImageNet21k","parameters":"303M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### De et al.

Accuracy: 96.6. Result ID: 699002963a6c314849fd727d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cbf1e209131cea35c41.md>)

[Implementation](<https://github.com/deepmind/jax_privacy>)

Recorded evaluation configuration: {"epsilon (ε)":"8","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### CrossViT base

Accuracy: 96.5. Result ID: 699002963a6c314849fd7281.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"8","pretrained\_dataset":"ImageNet1k","parameters":"104M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### CrossViT base

Accuracy: 96.2. Result ID: 699002963a6c314849fd7280.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"4","pretrained\_dataset":"ImageNet1k","parameters":"104M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### De et al.

Accuracy: 96.1. Result ID: 699002963a6c314849fd727c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cbf1e209131cea35c41.md>)

[Implementation](<https://github.com/deepmind/jax_privacy>)

Recorded evaluation configuration: {"epsilon (ε)":"4","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### CrossViT base

Accuracy: 96.1. Result ID: 699002963a6c314849fd727f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"2","pretrained\_dataset":"ImageNet1k","parameters":"104M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### CrossViT base

Accuracy: 95.5. Result ID: 699002963a6c314849fd727e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Implementation](<https://github.com/woodyx218/private_vision>)

Recorded evaluation configuration: {"epsilon (ε)":"1","pretrained\_dataset":"ImageNet1k","parameters":"104M","clipping\_method":"mixed ghost clipping"}. This record does not assert independent reproduction.

### De et al.

Accuracy: 95.4. Result ID: 699002963a6c314849fd727b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cbf1e209131cea35c41.md>)

[Implementation](<https://github.com/deepmind/jax_privacy>)

Recorded evaluation configuration: {"epsilon (ε)":"2","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### Yu et al.

Accuracy: 94.8. Result ID: 699002963a6c314849fd7278.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b59b.md>)

Recorded evaluation configuration: {"epsilon (ε)":"2","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### De et al.

Accuracy: 94.8. Result ID: 699002963a6c314849fd727a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cbf1e209131cea35c41.md>)

[Implementation](<https://github.com/deepmind/jax_privacy>)

Recorded evaluation configuration: {"epsilon (ε)":"1","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### Yu et al.

Accuracy: 94.3. Result ID: 699002963a6c314849fd7277.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b59b.md>)

Recorded evaluation configuration: {"epsilon (ε)":"1","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

### Tramer et al.

Accuracy: 92.7. Result ID: 699002963a6c314849fd7279.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e1b1e209131cea3dce4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa694b1e209131cea214b6.md>)

Recorded evaluation configuration: {"epsilon (ε)":"2","pretrained\_dataset":"ImageNet1k"}. This record does not assert independent reproduction.

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