RoI-level Breast Cancer Tumor Subtyping on BACH (test)
88.35AUROCGrapHist
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GrapHistAggregation=add-ABMIL2026.02 | 88.35 | — | |
| GrapHistAggregation=ABMIL2026.02 | 88.01 | — | |
| GrapHistAggregation=conj-ABMIL2026.02 | 86.47 | — | |
| DINOv2Aggregation=add-ABMIL2026.02 | 84.38 | — | |
| DINOv2Aggregation=ABMIL2026.02 | 80.87 | — | |
| MAEAggregation=add-ABMIL2026.02 | 80.76 | — | |
| MAEAggregation=conj-ABMIL2026.02 | 78.11 | — | |
| DINOv2Aggregation=conj-ABMIL2026.02 | 76.56 | — | |
| MAEAggregation=ABMIL2026.02 | 75.11 | — | |
| ACM-bioAggregation=add-ABMIL2026.02 | 66.44 | — | |
| ACM-UNIAggregation=add-ABMIL2026.02 | 64.9 | — | |
| ACM-UNIAggregation=ABMIL2026.02 | 55.97 | — | |
| ACM-UNIAggregation=conj-ABMIL2026.02 | 52.69 | — | |
| ACM-bioAggregation=ABMIL2026.02 | 50.21 | — | |
| ACM-bioAggregation=conj-ABMIL2026.02 | 46.73 | — | |
| ACM-bioModel Category=Supervised graph models, MIL aggregation=best variant2026.02 | — | 38.37 | |
| ACM-UNIModel Category=Supervised graph models, MIL aggregation=best variant2026.02 | — | 29.03 | |
| DINOv2Model Category=Self-supervised vision models, MIL aggregation=best variant2026.02 | — | 59.25 | |
| GrapHistModel Category=Ours, MIL aggregation=best variant2026.02 | — | 69.16 | |
| MAEModel Category=Self-supervised vision models, MIL aggregation=best variant2026.02 | — | 57.94 |