# DeepFake Detection on FF++ Adversarial Attack

Source: [DeepFake Detection on FF++ Adversarial Attack](<https://www.sota2.com/research/sota/deepfake-detection-on-ff-plus-plus-adversarial-attack>)

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

Data updated: 2026-02-26T05:31:56.979Z

Evaluation of DeepFake Detection performance on the FF++ dataset specifically configured for adversarial attack scenarios, reporting metrics such as Accuracy, F1-score, and 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 | DeepFake Detection |
| Dataset (including any stated split) | FF++ Adversarial Attack |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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/deepfake-detection-on-ff-plus-plus-adversarial-attack.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/deepfake-detection-on-ff-plus-plus-adversarial-attack.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [AUC](<https://www.sota2.com/research/sota/deepfake-detection-on-ff-plus-plus-adversarial-attack.md?metric=AUC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| LR-GCN | 91 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| Xception | 81.4 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| MaskRelation | 80.5 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| TALL-Swin | 79.6 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| RECCE | 79.4 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| CORE | 79.3 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| F3-net | 76.4 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| DFIL | 75.5 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| UCF | 72.7 | Not recorded | 6937ad5f6cfd488761a4ed68 |
| DFGaze | 70.7 | Not recorded | 6937ad5f6cfd488761a4ed68 |

### LR-GCN

Accuracy: 91. Result ID: 699876675698665ec4d6d6a3.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### Xception

Accuracy: 81.4. Result ID: 699876675698665ec4d6d69a.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63371e209131ce9fd796.md>)

[Implementation](<https://keras.io/applications/#xception>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### MaskRelation

Accuracy: 80.5. Result ID: 699876675698665ec4d6d6a2.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### TALL-Swin

Accuracy: 79.6. Result ID: 699876675698665ec4d6d6a0.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72381e209131cea55071.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### RECCE

Accuracy: 79.4. Result ID: 699876675698665ec4d6d69c.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### CORE

Accuracy: 79.3. Result ID: 699876675698665ec4d6d69d.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb31e209131cea35555.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### F3-net

Accuracy: 76.4. Result ID: 699876675698665ec4d6d69b.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa689c1e209131cea1d080.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### DFIL

Accuracy: 75.5. Result ID: 699876675698665ec4d6d69f.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa709d1e209131cea4c408.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### UCF

Accuracy: 72.7. Result ID: 699876675698665ec4d6d69e.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71021e209131cea4e7e8.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

### DFGaze

Accuracy: 70.7. Result ID: 699876675698665ec4d6d6a1.

[Reporting paper](<https://www.sota2.com/research/paper/6937ad5f6cfd488761a4ed68.md>)

Recorded evaluation configuration: {"mask ratio (mr)":0.2}. This record does not assert independent reproduction.

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