# Image Classification on ImageNet (Clean Accuracy, Robust Accuracy)

Source: [Image Classification on ImageNet (Clean Accuracy, Robust Accuracy)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-clean-accuracy-robust-accuracy>)

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

Data updated: 2026-02-27T11:16:50.187Z

Evaluation of Image Classification performance on the standard ImageNet dataset, focusing on robustness metrics such as Clean Accuracy and Robust Accuracy.

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 |
| Metric | Clean Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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-clean-accuracy-robust-accuracy.md?metric=Clean%20Accuracy>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Clean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| TTE | 71.5 | Not recorded | 69799806a2b28e761d87cb52 |
| R-TPT | 71.2 | Not recorded | 69799806a2b28e761d87cb52 |
| CLIP | 69.9 | Not recorded | 69799806a2b28e761d87cb52 |
| TeCoA | 69.2 | Not recorded | 69799806a2b28e761d87cb52 |
| Anti-Adv | 68 | Not recorded | 69799806a2b28e761d87cb52 |
| CSR | 68 | Not recorded | 69799806a2b28e761d87cb52 |
| HD | 66.4 | Not recorded | 69799806a2b28e761d87cb52 |
| LPF | 64.2 | Not recorded | 69799806a2b28e761d87cb52 |
| FARE | 63.9 | Not recorded | 69799806a2b28e761d87cb52 |
| TTC | 54.7 | Not recorded | 69799806a2b28e761d87cb52 |

### TTE

Clean Accuracy: 71.5. Result ID: 69a17d1735dfcabb00e7627d.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a711e209131cea2817e.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### R-TPT

Clean Accuracy: 71.2. Result ID: 69a17d1735dfcabb00e76279.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7add1e209131cea8590d.md>)

[Implementation](<https://github.com/TomSheng21/R-TPT>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### CLIP

Clean Accuracy: 69.9. Result ID: 69a17d1735dfcabb00e76276.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

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

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

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Original"}. This record does not assert independent reproduction.

### TeCoA

Clean Accuracy: 69.2. Result ID: 69a17d1735dfcabb00e76277.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

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

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Adversarial Fine-tuning"}. This record does not assert independent reproduction.

### Anti-Adv

Clean Accuracy: 68. Result ID: 69a17d1735dfcabb00e7627c.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b7f1e209131cea2e9af.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### CSR

Clean Accuracy: 68. Result ID: 69a17d1735dfcabb00e7627f.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### HD

Clean Accuracy: 66.4. Result ID: 69a17d1735dfcabb00e7627b.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a1d1e209131cea26045.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### LPF

Clean Accuracy: 64.2. Result ID: 69a17d1735dfcabb00e7627a.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

### FARE

Clean Accuracy: 63.9. Result ID: 69a17d1735dfcabb00e76278.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

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

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Adversarial Fine-tuning"}. This record does not assert independent reproduction.

### TTC

Clean Accuracy: 54.7. Result ID: 69a17d1735dfcabb00e7627e.

[Reporting paper](<https://www.sota2.com/research/paper/69799806a2b28e761d87cb52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa774a1e209131cea724cc.md>)

[Implementation](<https://www.sota2.com/here>)

Recorded evaluation configuration: {"Backbone":"CLIP-L/14","Method category":"Test-Time Defense"}. This record does not assert independent reproduction.

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