BAP1 Mutation Prediction on BAP1 mutation (held-out)
78Balanced AccuracyMOOZY
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MOOZYEvaluation Protocol=Frozen model evaluated with MLP probe2026.03 | 78 | 89 | 79 | — | — | |
| MOOZYEvaluation Protocol=Frozen-feature MLP probe2026.03 | 78 | — | — | 89 | 79 | |
| CHIEFEvaluation Protocol=Frozen-feature MLP probe2026.03 | 75 | — | — | 86 | 75 | |
| MadeleineEvaluation Protocol=Frozen-feature MLP probe2026.03 | 75 | — | — | 85 | 78 | |
| TITANEvaluation Protocol=Frozen-feature MLP probe2026.03 | 75 | — | — | 84 | 82 | |
| CONCH v1.5Evaluation Protocol=Aggregator trained from scratch on frozen features, Aggregator=Mean over five MIL architectures (MeanMIL, ABMIL, CLAM, DSMIL, TransMIL)2026.03 | 74 | 85 | 75 | — | — | |
| MUSKEvaluation Protocol=Aggregator trained from scratch on frozen features, Aggregator=Mean over five MIL architectures (MeanMIL, ABMIL, CLAM, DSMIL, TransMIL)2026.03 | 73 | 86 | 73 | — | — | |
| Phikon v2Evaluation Protocol=Aggregator trained from scratch on frozen features, Aggregator=Mean over five MIL architectures (MeanMIL, ABMIL, CLAM, DSMIL, TransMIL)2026.03 | 68 | 83 | 67 | — | — | |
| BackboneEvaluation Protocol=Aggregator trained from scratch on frozen features, Aggregator=Mean over five MIL architectures (MeanMIL, ABMIL, CLAM, DSMIL, TransMIL)2026.03 | 67 | 84 | 66 | — | — | |
| Giga PathEvaluation Protocol=Frozen-feature MLP probe2026.03 | 66 | — | — | 84 | 63 | |
| PRISMEvaluation Protocol=Frozen-feature MLP probe2026.03 | 66 | — | — | 80 | 71 | |
| UNI v2Evaluation Protocol=Aggregator trained from scratch on frozen features, Aggregator=Mean over five MIL architectures (MeanMIL, ABMIL, CLAM, DSMIL, TransMIL)2026.03 | 64 | 82 | 65 | — | — |