# Image Classification on ImageNet-1K (Robust Accuracy PGD-100)

Source: [Image Classification on ImageNet-1K (Robust Accuracy PGD-100)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-robust-accuracy-pgd-100>)

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

Data updated: 2026-02-27T13:16:43.478Z

Evaluation of Image Classification performance on the ImageNet-1K dataset, reporting clean accuracy alongside robustness against adversarial attacks (PGD-100).

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-1K |
| Metric | Clean Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Clean Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-robust-accuracy-pgd-100.md?metric=Clean%20Accuracy>): Higher is better; unit: PERCENT.

- [Robust Accuracy (PGD-100, eps=1/255)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-robust-accuracy-pgd-100.md?metric=Robust%20Accuracy%20(PGD-100%2C%20eps%3D1%2F255)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Clean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FARE | 51.78 | Not recorded | 6992a4ad9ebc499c97406d32 |
| SAFT-M | 47.3 | Not recorded | 6992a4ad9ebc499c97406d32 |
| SAFT-L | 46.61 | Not recorded | 6992a4ad9ebc499c97406d32 |
| TGA-ZSR | 46.5 | Not recorded | 6992a4ad9ebc499c97406d32 |
| CLIP | 44.4 | Not recorded | 6992a4ad9ebc499c97406d32 |
| TeCoA | 31.85 | Not recorded | 6992a4ad9ebc499c97406d32 |
| PMG-AFT | 23.72 | Not recorded | 6992a4ad9ebc499c97406d32 |

### FARE

Clean Accuracy: 51.78. Result ID: 69a1992435dfcabb00e7dcd1.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d24a.md>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### SAFT-M

Clean Accuracy: 47.3. Result ID: 69a1992435dfcabb00e7dcd5.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### SAFT-L

Clean Accuracy: 46.61. Result ID: 69a1992435dfcabb00e7dcd4.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### TGA-ZSR

Clean Accuracy: 46.5. Result ID: 69a1992435dfcabb00e7dcd3.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75771e209131cea68256.md>)

[Implementation](<https://github.com/zhyblue424/TGA-ZSR>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### CLIP

Clean Accuracy: 44.4. Result ID: 69a1992435dfcabb00e7dccf.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### TeCoA

Clean Accuracy: 31.85. Result ID: 69a1992435dfcabb00e7dcd0.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f271e209131cea44177.md>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### PMG-AFT

Clean Accuracy: 23.72. Result ID: 69a1992435dfcabb00e7dcd2.

[Reporting paper](<https://www.sota2.com/research/paper/6992a4ad9ebc499c97406d32.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72df1e209131cea5912e.md>)

Recorded evaluation configuration: {"Zero-shot evaluation":"true","Adversarial Attack":"PGD-100 (epsilon = 1/255)","Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

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