MLAAD
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Audio Deepfake Detection | MLAAD-EN | EER10.7 | 18 | |
| Global Source Attribution | MLAAD Few-Shot Open-Set (test) | F1-Macro (All vs All)84.52 | 16 | |
| Global Source Attribution | MLAAD Closed-Set IID (test) | F1-Macro (Seen IID)95.44 | 13 | |
| Audio Deepfake Detection | MLAAD v1 | Accuracy71.8 | 12 | |
| Audio Source Attribution | MLAAD (OOD Eval) | Accuracy94.3 | 11 | |
| Audio Source Attribution | MLAAD In-Domain (Eval) | Accuracy97.6 | 11 | |
| Audio Deepfake Detection | MLAAD (eval) | EER (%)10.9 | 6 | |
| Audio Source Tracing | MLAAD (eval) | ID Accuracy97.6 | 5 | |
| Source verification | MLAAD open-set in-domain (test) | EER10.5 | 4 | |
| Closed-set Attribution | MLAAD v5 | Accuracy98.57 | 3 | |
| OOD Detection | MLAAD v5 | Accuracy89.2 | 2 | |
| Audio Deepfake Detection | MLAAD (test) | Accuracy80 | 2 |
Showing 12 of 12 rows