Uncertainty Estimation on OBQA
88.03AUROCDeepEns
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepEnsBackbone=Gemma-2-9B, Measure=MP2026.04 | 88.03 | |
| ETNBackbone=Llama-3.1-8B, Measure=MP2026.04 | 83.21 | |
| MCDBackbone=Llama-3.1-8B, Measure=MP2026.04 | 83.02 | |
| ETNBackbone=Gemma-2-9B, Measure=MP2026.04 | 82.68 | |
| MAP_CEBackbone=Llama-3.1-8B, Measure=MP2026.04 | 81.94 | |
| MCDBackbone=Gemma-2-9B, Measure=MP2026.04 | 81.83 | |
| MAP_CEBackbone=Gemma-2-9B, Measure=MP2026.04 | 81.35 | |
| IB-EDLBackbone=Gemma-2-9B, Measure=MP2026.04 | 80.67 | |
| DeepEnsBackbone=Llama-3.1-8B, Measure=MP2026.04 | 80.61 | |
| IB-EDLBackbone=Llama-3.1-8B, Measure=MP2026.04 | 79.98 | |
| DMMBackbone=Gemma-2-9B, Measure=MP2026.04 | 77.97 | |
| DMMBackbone=Llama-3.1-8B, Measure=MP2026.04 | 77.12 | |
| MAP_EDLBackbone=Gemma-2-9B, Measure=MP2026.04 | 76.9 | |
| DMMBackbone=Gemma-2-9B, Measure=UM2026.04 | 75.12 | |
| ETNBackbone=Llama-3.1-8B, Measure=UM2026.04 | 73.6 | |
| DMMBackbone=Llama-3.1-8B, Measure=UM2026.04 | 69.32 | |
| IB-EDLBackbone=Gemma-2-9B, Measure=UM2026.04 | 69.3 | |
| ETNBackbone=Gemma-2-9B, Measure=UM2026.04 | 64.87 | |
| MAP_EDLBackbone=Llama-3.1-8B, Measure=MP2026.04 | 64.44 | |
| IB-EDLBackbone=Llama-3.1-8B, Measure=UM2026.04 | 63.71 | |
| MAP_EDLBackbone=Gemma-2-9B, Measure=UM2026.04 | 56.79 | |
| MAP_EDLBackbone=Llama-3.1-8B, Measure=UM2026.04 | 48.6 | |
| LLBackbone=Llama-3.1-8B, Measure=MP2026.04 | 41.32 | |
| LLBackbone=Gemma-2-9B, Measure=MP2026.04 | 29.94 |