Abnormality Detection on TUAB
83.2Balanced AccuracyKAST-BAR-Large
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| KAST-BAR-LargeModel Parameter=2.2B, General Model=true, Multi-Task=true2026.05 | 83.2 | 87.1 | 90.7 | |
| THD-BAR-HugeModel Parameter=1.6B, General Model=true, Multi-Task=true2026.05 | 82.2 | 84.8 | 88.6 | |
| CSBrainModel Parameter=4.9M, General Model=true, Multi-Task=false2026.05 | 81.7 | 90.1 | 89.6 | |
| TFM-TokenizerModel Size=1.9M, Training Setting=Single Dataset Setting2025.02 | 81.52 | 89.46 | 88.97 | |
| LaBraM-BaseModel Parameter=5.8M, General Model=true, Multi-Task=false2026.05 | 81.4 | 89.7 | 90.2 | |
| KAST-BAR-BaseModel Parameter=0.8B, General Model=true, Multi-Task=true2026.05 | 81.3 | 85.6 | 88.2 | |
| TFM-Tokenizer†Model Size=1.9M, Training Setting=Multiple Dataset Pretraining, Pretraining_Scope=4 EEG datasets2025.02 | 80.32 | 88.86 | 88.7 | |
| NeuroLM-XLModel Parameter=1.7B, General Model=true, Multi-Task=true2026.05 | 79.7 | 72.2 | 78.8 | |
| ST-TransformerModel Size=3.5M, Training Setting=Single Dataset Setting2025.02 | 79.66 | 85.21 | 87.07 | |
| BIOTModel Parameter=3.2M, General Model=true, Multi-Task=false2026.05 | 79.6 | 87.9 | 88.2 | |
| BIOTModel Size=3.2M, Training Setting=Multiple Dataset Pretraining2025.02 | 79.59 | 87.92 | 88.15 | |
| EEGPTModel Size=4.7M, Training Setting=Multiple Dataset Pretraining2025.02 | 79.59 | — | 87.16 | |
| BIOT*Model Size=3.2M, Training Setting=Single Dataset Setting, Reproduced=true2025.02 | 79.55 | 88.19 | 88.34 | |
| ST-TransformerModel Parameter=3.5M, General Model=false, Multi-Task=false2026.05 | 79.3 | 85.4 | 86.9 | |
| Vanilla BIOTModel Size=3.2M, Training Setting=Single Dataset Setting2025.02 | 79.25 | 87.07 | 86.91 | |
| EEGPTModel Parameter=25M, General Model=true, Multi-Task=true2026.05 | 79.2 | 74.6 | 86.6 | |
| SPaRCNetModel Size=0.79M, Training Setting=Single Dataset Setting2025.02 | 78.96 | 84.14 | 86.76 | |
| CBraModModel Parameter=4.0M, General Model=true, Multi-Task=false2026.05 | 78.9 | 86.4 | 86.1 | |
| FFCLModel Size=2.4M, Training Setting=Single Dataset Setting2025.02 | 78.48 | 84.48 | 85.69 | |
| NeuroLM-BModel Size=254M, Training Setting=Multiple Dataset Pretraining2025.02 | 78.26 | 69.75 | 78.16 | |
| CNN-TransformerModel Size=3.2M, Training Setting=Single Dataset Setting2025.02 | 77.77 | 84.33 | 84.61 | |
| SPaRCNetModel Parameter=0.79M, General Model=false, Multi-Task=false2026.05 | 77.5 | 83.1 | 86.3 | |
| ContraWRModel Size=1.6M, Training Setting=Single Dataset Setting2025.02 | 77.46 | 84.21 | 84.56 | |
| LaBraM-Base†Model Size=5.8M, Training Setting=Multiple Dataset Pretraining, Pretraining_Scope=4 EEG datasets2025.02 | 77.35 | 85.31 | 85.57 | |
| LaBraM-Base*Model Size=5.8M, Training Setting=Single Dataset Setting, Reproduced=true2025.02 | 77.2 | 84.98 | 85.34 | |
| EEGNetGeneral Model=false, Multi-Task=false2026.05 | 77.1 | 82.3 | 85 | |
| CBraMod†Model Size=4M, Training Setting=Multiple Dataset Pretraining, Pretraining_Scope=4 EEG datasets2025.02 | 50 | 49.38 | 52.81 |