# Image Classification on ImageNet Robustness Suite (IN-C, C-bar, A, D, Sketch, Drawing, Cartoon, Hard) (val)

Source: [Image Classification on ImageNet Robustness Suite (IN-C, C-bar, A, D, Sketch, Drawing, Cartoon, Hard) (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val>)

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

Data updated: 2026-02-26T05:44:08.859Z

Evaluation of Image Classification performance across specific robustness subsets of ImageNet, including IN-C, C-bar, A, D, Sketch, Drawing, Cartoon, and Hard variants, measured on the validation split.

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 Robustness Suite IN-C, C-bar, A, D, Sketch, Drawing, Cartoon, Hard (val) |
| Metric | FGSM 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

- [FGSM Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=FGSM%20Accuracy>): Higher is better; unit: PERCENT.

- [Robustness A Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Robustness%20A%20Accuracy>): Higher is better; unit: PERCENT.

- [ImageNet-C Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=ImageNet-C%20Accuracy>): Higher is better; unit: PERCENT.

- [ImageNet-C-bar Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=ImageNet-C-bar%20Accuracy>): Higher is better; unit: PERCENT.

- [Cartoon Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Cartoon%20Accuracy>): Higher is better; unit: PERCENT.

- [Robustness D Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Robustness%20D%20Accuracy>): Higher is better; unit: PERCENT.

- [Drawing Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Drawing%20Accuracy>): Higher is better; unit: PERCENT.

- [Sketch Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Sketch%20Accuracy>): Higher is better; unit: PERCENT.

- [Hard Corruption Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Hard%20Corruption%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-robustness-suite-in-c-c-bar-a-d-sketch-drawing-cartoon-hard-val.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | FGSM Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Dropout | 62.45 | Not recorded | 68fa765f1e209131cea6cb82 |
| SAM | 61.85 | Not recorded | 68fa765f1e209131cea6cb82 |
| SAM | 59.92 | Not recorded | 68fa765f1e209131cea6cb82 |
| DAMP | 59.87 | Not recorded | 68fa765f1e209131cea6cb82 |
| ASAM | 59.38 | Not recorded | 68fa765f1e209131cea6cb82 |
| ASAM | 58.87 | Not recorded | 68fa765f1e209131cea6cb82 |
| Dropout | 56.43 | Not recorded | 68fa765f1e209131cea6cb82 |
| DAMP | 55.2 | Not recorded | 68fa765f1e209131cea6cb82 |

### Dropout

FGSM Accuracy: 62.45. Result ID: 69941c8e3a6c3148491b22bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

Recorded evaluation configuration: {"Model":"ViT-S16","#Epochs":500,"Runtime":"111h"}. This record does not assert independent reproduction.

### SAM

FGSM Accuracy: 61.85. Result ID: 69941c8e3a6c3148491b22c3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69d81e209131cea248fb.md>)

[Implementation](<https://github.com/google-research/sam>)

Recorded evaluation configuration: {"Model":"ViT-B16","#Epochs":150,"Runtime":"135h"}. This record does not assert independent reproduction.

### SAM

FGSM Accuracy: 59.92. Result ID: 69941c8e3a6c3148491b22bf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69d81e209131cea248fb.md>)

[Implementation](<https://github.com/google-research/sam>)

Recorded evaluation configuration: {"Model":"ViT-S16","#Epochs":300,"Runtime":"123h"}. This record does not assert independent reproduction.

### DAMP

FGSM Accuracy: 59.87. Result ID: 69941c8e3a6c3148491b22be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

Recorded evaluation configuration: {"Model":"ViT-S16","#Epochs":500,"Runtime":"111h"}. This record does not assert independent reproduction.

### ASAM

FGSM Accuracy: 59.38. Result ID: 69941c8e3a6c3148491b22c0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a441e209131cea27274.md>)

Recorded evaluation configuration: {"Model":"ViT-S16","#Epochs":300,"Runtime":"123h"}. This record does not assert independent reproduction.

### ASAM

FGSM Accuracy: 58.87. Result ID: 69941c8e3a6c3148491b22c4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a441e209131cea27274.md>)

Recorded evaluation configuration: {"Model":"ViT-B16","#Epochs":150,"Runtime":"136h"}. This record does not assert independent reproduction.

### Dropout

FGSM Accuracy: 56.43. Result ID: 69941c8e3a6c3148491b22c1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

Recorded evaluation configuration: {"Model":"ViT-B16","#Epochs":275,"Runtime":"123h"}. This record does not assert independent reproduction.

### DAMP

FGSM Accuracy: 55.2. Result ID: 69941c8e3a6c3148491b22c2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb82.md>)

Recorded evaluation configuration: {"Model":"ViT-B16","#Epochs":275,"Runtime":"124h"}. This record does not assert independent reproduction.

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