Uncertainty Estimation on Shared (GQA, POPE, etc.) (test)
0.001ECEPIK
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| PIKAggregation Protocol=per-dataset equal-weight, Comparison=Significance vs BICR, Baseline Type=Trained2026.05 | 0.001 | 0.001 | — | — | |
| SAPLMAAggregation Protocol=per-dataset equal-weight, Comparison=Significance vs BICR, Baseline Type=Trained2026.05 | 0.001 | 0.001 | 0.1 | 0.1 | |
| CCPSAggregation Protocol=per-dataset equal-weight, Comparison=Significance vs BICR, Baseline Type=Trained2026.05 | 0.001 | 0.001 | 0.1 | 0.1 | |
| IIAggregation Protocol=per-dataset equal-weight, Comparison=Significance vs BICR, Baseline Type=Trained, Full Name=InternalInspector2026.05 | 0.007 | 0.001 | 0.1 | 0.1 |