Uncertainty Calibration on GPQA diamond
0.696AUROCET-PE
Evaluation Results
| Method | Links | |
|---|---|---|
| ET-PEBackbone=Qwen2.5-14B-Instruct2026.01 | 0.696 | |
| ET-PEBackbone=Qwen2.5-32B-Instruct2026.01 | 0.684 | |
| LSBackbone=Qwen2.5-32B-Instruct2026.01 | 0.682 | |
| PEBackbone=Qwen2.5-14B-Instruct2026.01 | 0.671 | |
| PEBackbone=Qwen2.5-32B-Instruct2026.01 | 0.649 | |
| ET-PEBackbone=Qwen2.5-7B-Instruct2026.01 | 0.638 | |
| p(True)Backbone=Qwen2.5-7B-Instruct2026.01 | 0.638 | |
| LSBackbone=Qwen2.5-7B-Instruct2026.01 | 0.631 | |
| p(True)Backbone=Qwen2.5-32B-Instruct2026.01 | 0.627 | |
| p(True)Backbone=Qwen2.5-14B-Instruct2026.01 | 0.626 | |
| PEBackbone=Qwen2.5-7B-Instruct2026.01 | 0.621 | |
| LSBackbone=Qwen2.5-14B-Instruct2026.01 | 0.607 | |
| SEBackbone=Qwen2.5-32B-Instruct2026.01 | 0.604 | |
| LN-PEBackbone=Qwen2.5-32B-Instruct2026.01 | 0.583 | |
| SEBackbone=Qwen2.5-14B-Instruct2026.01 | 0.582 | |
| LN-PEBackbone=Qwen2.5-14B-Instruct2026.01 | 0.578 | |
| SEBackbone=Qwen2.5-7B-Instruct2026.01 | 0.484 | |
| LN-PEBackbone=Qwen2.5-7B-Instruct2026.01 | 0.46 |