# Perturbation Test on CIFAR-100 1.0 (test)

Source: [Perturbation Test on CIFAR-100 1.0 (test)](<https://www.sota2.com/research/sota/perturbation-test-on-cifar-100-1-0-test>)

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

Data updated: 2026-02-26T05:25:22.052Z

Evaluation of model sensitivity to input perturbations on the CIFAR-100 test set, measured using AUC scores derived from predicted and ground truth labels.

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 | Perturbation Test |
| Dataset (including any stated split) | CIFAR-100 1.0 (test) |
| Metric | AUC (Predicted, Neg) |
| Unit | SCALAR |
| 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

- [AUC (Predicted, Neg)](<https://www.sota2.com/research/sota/perturbation-test-on-cifar-100-1-0-test.md?metric=AUC%20(Predicted%2C%20Neg)>): Higher is better; unit: SCALAR.

- [AUC (Predicted, Pos)](<https://www.sota2.com/research/sota/perturbation-test-on-cifar-100-1-0-test.md?metric=AUC%20(Predicted%2C%20Pos)>): Higher is better; unit: SCALAR.

- [AUC (GT, Neg)](<https://www.sota2.com/research/sota/perturbation-test-on-cifar-100-1-0-test.md?metric=AUC%20(GT%2C%20Neg)>): Higher is better; unit: SCALAR.

- [AUC (GT, Pos)](<https://www.sota2.com/research/sota/perturbation-test-on-cifar-100-1-0-test.md?metric=AUC%20(GT%2C%20Pos)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUC (Predicted, Neg) | Result date | Report paper |
| --- | --- | --- | --- |
| TokenTM | 0.5478 | Not recorded | 68fa729e1e209131cea57935 |
| GAE | 0.5214 | Not recorded | 68fa729e1e209131cea57935 |
| Grad-CAM | 0.4718 | Not recorded | 68fa729e1e209131cea57935 |
| LRP | 0.4676 | Not recorded | 68fa729e1e209131cea57935 |
| Trans. Attr. | 0.4465 | Not recorded | 68fa729e1e209131cea57935 |
| Rollout | 0.4134 | Not recorded | 68fa729e1e209131cea57935 |
| Raw Att. | 0.3723 | Not recorded | 68fa729e1e209131cea57935 |
| Cons. LRP | 0.2711 | Not recorded | 68fa729e1e209131cea57935 |
| ATTCAT | 0.2352 | Not recorded | 68fa729e1e209131cea57935 |

### TokenTM

AUC (Predicted, Neg): 0.5478. Result ID: 6992643c3a6c3148490ea292.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### GAE

AUC (Predicted, Neg): 0.5214. Result ID: 6992643c3a6c3148490ea291.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea233a8.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### Grad-CAM

AUC (Predicted, Neg): 0.4718. Result ID: 6992643c3a6c3148490ea28a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12568.md>)

[Implementation](<https://github.com/ramprs/grad-cam/>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### LRP

AUC (Predicted, Neg): 0.4676. Result ID: 6992643c3a6c3148490ea28b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa625f1e209131ce9f8e34.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### Trans. Attr.

AUC (Predicted, Neg): 0.4465. Result ID: 6992643c3a6c3148490ea28d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b41e209131cea2394f.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### Rollout

AUC (Predicted, Neg): 0.4134. Result ID: 6992643c3a6c3148490ea28f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67ae1e209131cea17be6.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### Raw Att.

AUC (Predicted, Neg): 0.3723. Result ID: 6992643c3a6c3148490ea28e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa665a1e209131cea0fbee.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### Cons. LRP

AUC (Predicted, Neg): 0.2711. Result ID: 6992643c3a6c3148490ea28c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd51e209131cea3613b.md>)

[Implementation](<https://github.com/AmeenAli/XAI_Transformers>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

### ATTCAT

AUC (Predicted, Neg): 0.2352. Result ID: 6992643c3a6c3148490ea290.

[Reporting paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57935.md>)

Recorded evaluation configuration: {"Backbone":"DeiT"}. This record does not assert independent reproduction.

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