# Adversarial Robustness on ImageNet-LT

Source: [Adversarial Robustness on ImageNet-LT](<https://www.sota2.com/research/sota/adversarial-robustness-on-imagenet-lt>)

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

Data updated: 2026-02-26T05:39:19.021Z

Evaluation of model performance on the ImageNet-LT dataset under adversarial attack scenarios, typically measured by clean accuracy and accuracy under FGSM and PGD attacks.

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 | Adversarial Robustness |
| Dataset (including any stated split) | ImageNet-LT |
| Metric | Clean Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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/adversarial-robustness-on-imagenet-lt.md?metric=Clean%20Accuracy>): Higher is better; unit: PERCENT.

- [FGSM Accuracy](<https://www.sota2.com/research/sota/adversarial-robustness-on-imagenet-lt.md?metric=FGSM%20Accuracy>): Higher is better; unit: PERCENT.

- [PGD-20 Accuracy](<https://www.sota2.com/research/sota/adversarial-robustness-on-imagenet-lt.md?metric=PGD-20%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Clean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FGSM-Our | 30.02 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FGSM-Our | 25.88 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FGSM-AT | 25.64 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FGSM-AT | 21.94 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FAST-Our | 15.45 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FAST-AT | 11.36 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FAST-Our | 10.76 | Not recorded | 68fa6a8c1e209131cea28b65 |
| FAST-AT | 7.2 | Not recorded | 68fa6a8c1e209131cea28b65 |

### FGSM-Our

Clean Accuracy: 30.02. Result ID: 699ab8405698665ec446327a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Implementation](<https://github.com/wutong16/Adversarial_Long-Tail>)

Recorded evaluation configuration: {"ε":"2/255"}. This record does not assert independent reproduction.

### FGSM-Our

Clean Accuracy: 25.88. Result ID: 699ab8405698665ec446327e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Implementation](<https://github.com/wutong16/Adversarial_Long-Tail>)

Recorded evaluation configuration: {"ε":"4/255"}. This record does not assert independent reproduction.

### FGSM-AT

Clean Accuracy: 25.64. Result ID: 699ab8405698665ec4463279.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a3f.md>)

[Implementation](<https://github.com/ashafahi/free_adv_train>)

Recorded evaluation configuration: {"ε":"2/255"}. This record does not assert independent reproduction.

### FGSM-AT

Clean Accuracy: 21.94. Result ID: 699ab8405698665ec446327d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a3f.md>)

[Implementation](<https://github.com/ashafahi/free_adv_train>)

Recorded evaluation configuration: {"ε":"4/255"}. This record does not assert independent reproduction.

### FAST-Our

Clean Accuracy: 15.45. Result ID: 699ab8405698665ec4463278.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Implementation](<https://github.com/wutong16/Adversarial_Long-Tail>)

Recorded evaluation configuration: {"ε":"2/255"}. This record does not assert independent reproduction.

### FAST-AT

Clean Accuracy: 11.36. Result ID: 699ab8405698665ec4463277.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66e51e209131cea12efc.md>)

Recorded evaluation configuration: {"ε":"2/255"}. This record does not assert independent reproduction.

### FAST-Our

Clean Accuracy: 10.76. Result ID: 699ab8405698665ec446327c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Implementation](<https://github.com/wutong16/Adversarial_Long-Tail>)

Recorded evaluation configuration: {"ε":"4/255"}. This record does not assert independent reproduction.

### FAST-AT

Clean Accuracy: 7.2. Result ID: 699ab8405698665ec446327b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b65.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66e51e209131cea12efc.md>)

Recorded evaluation configuration: {"ε":"4/255"}. This record does not assert independent reproduction.

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