# Trojaned Model Detection on MNIST Resnet18 (test)

Source: [Trojaned Model Detection on MNIST Resnet18 (test)](<https://www.sota2.com/research/sota/trojaned-model-detection-on-mnist-resnet18-test>)

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

Data updated: 2026-02-26T05:27:55.538Z

Evaluation of a model's capability to detect Trojan attacks specifically on the test split of a ResNet18 model trained on MNIST, measured by Accuracy (ACC) and Area Under the Curve (AUC).

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 | Trojaned Model Detection |
| Dataset (including any stated split) | MNIST Resnet18 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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/trojaned-model-detection-on-mnist-resnet18-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUC](<https://www.sota2.com/research/sota/trojaned-model-detection-on-mnist-resnet18-test.md?metric=AUC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Topo | 87 | Not recorded | 68fa6a211e209131cea2628f |
| ULP | 71 | Not recorded | 68fa6a211e209131cea2628f |
| NC | 65 | Not recorded | 68fa6a211e209131cea2628f |
| Corr | 56 | Not recorded | 68fa6a211e209131cea2628f |
| DFTND | 53 | Not recorded | 68fa6a211e209131cea2628f |

### Topo

Accuracy: 87. Result ID: 698e1237ffc3901de09a9237.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

Recorded evaluation configuration: {"architecture":"Resnet18"}. This record does not assert independent reproduction.

### ULP

Accuracy: 71. Result ID: 698e1237ffc3901de09a9235.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67881e209131cea16b47.md>)

Recorded evaluation configuration: {"architecture":"Resnet18"}. This record does not assert independent reproduction.

### NC

Accuracy: 65. Result ID: 698e1237ffc3901de09a9233.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

Recorded evaluation configuration: {"architecture":"Resnet18"}. This record does not assert independent reproduction.

### Corr

Accuracy: 56. Result ID: 698e1237ffc3901de09a9236.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

Recorded evaluation configuration: {"architecture":"Resnet18"}. This record does not assert independent reproduction.

### DFTND

Accuracy: 53. Result ID: 698e1237ffc3901de09a9234.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a211e209131cea2628f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680b1e209131cea19c39.md>)

Recorded evaluation configuration: {"architecture":"Resnet18"}. This record does not assert independent reproduction.

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