# Image Classification on ImageNet-9 adversarial backgrounds

Source: [Image Classification on ImageNet-9 adversarial backgrounds](<https://www.sota2.com/research/sota/image-classification-on-imagenet-9-adversarial-backgrounds>)

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

Data updated: 2026-02-26T05:28:17.042Z

Evaluation of Image Classification performance on the ImageNet-9 dataset specifically configured with adversarial backgrounds.

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-9 adversarial backgrounds |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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-imagenet-9-adversarial-backgrounds.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Causal Transportability | 0.2941 | Not recorded | 68fa6c681e209131cea33e46 |
| Ablation | 0.2844 | Not recorded | 68fa6c681e209131cea33e46 |
| ERM | 0.2773 | Not recorded | 68fa6c681e209131cea33e46 |
| Causal Transportability | 0.2042 | Not recorded | 68fa6c681e209131cea33e46 |
| Ablation | 0.2025 | Not recorded | 68fa6c681e209131cea33e46 |
| ERM | 0.2 | Not recorded | 68fa6c681e209131cea33e46 |
| Causal Transportability | 0.1802 | Not recorded | 68fa6c681e209131cea33e46 |
| Ablation | 0.1704 | Not recorded | 68fa6c681e209131cea33e46 |
| ERM | 0.1459 | Not recorded | 68fa6c681e209131cea33e46 |

### Causal Transportability

Accuracy: 0.2941. Result ID: 698f11a83a6c314849f576ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"SimCLR"}. This record does not assert independent reproduction.

### Ablation

Accuracy: 0.2844. Result ID: 698f11a83a6c314849f576b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"SimCLR"}. This record does not assert independent reproduction.

### ERM

Accuracy: 0.2773. Result ID: 698f11a83a6c314849f576b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

Recorded evaluation configuration: {"Pre-trained representation":"SimCLR"}. This record does not assert independent reproduction.

### Causal Transportability

Accuracy: 0.2042. Result ID: 698f11a83a6c314849f576b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"SWAV"}. This record does not assert independent reproduction.

### Ablation

Accuracy: 0.2025. Result ID: 698f11a83a6c314849f576b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"SWAV"}. This record does not assert independent reproduction.

### ERM

Accuracy: 0.2. Result ID: 698f11a83a6c314849f576b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

Recorded evaluation configuration: {"Pre-trained representation":"SWAV"}. This record does not assert independent reproduction.

### Causal Transportability

Accuracy: 0.1802. Result ID: 698f11a83a6c314849f576b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"Moco-v2"}. This record does not assert independent reproduction.

### Ablation

Accuracy: 0.1704. Result ID: 698f11a83a6c314849f576b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

[Implementation](<https://github.com/cvlab-columbia/CT4Recognition>)

Recorded evaluation configuration: {"Pre-trained representation":"Moco-v2"}. This record does not assert independent reproduction.

### ERM

Accuracy: 0.1459. Result ID: 698f11a83a6c314849f576b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c681e209131cea33e46.md>)

Recorded evaluation configuration: {"Pre-trained representation":"Moco-v2"}. This record does not assert independent reproduction.

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