Expected Calibration Error on BT-guided
2ECE (15-bin)PPT (Distribution)
Evaluation Results
| Method | Links | |
|---|---|---|
| PPT (Distribution)Model=LLAMA3-8B, Temperature Scaling=true2026.05 | 2 | |
| PPT (Distribution)Model=QWEN2-7B, Temperature Scaling=true2026.05 | 5.1 | |
| PPT (Distribution)Model=LLAMA3-8B, Temperature Scaling=false2026.05 | 13.4 | |
| PPT (Distribution)Model=QWEN2-7B, Temperature Scaling=false2026.05 | 22.8 | |
| PPT (Mean)Model=LLAMA3-8B, Temperature Scaling=false2026.05 | 24.6 | |
| VerbalizedModel=LLAMA3-8B, Temperature Scaling=false2026.05 | 31.2 | |
| PPT (Mean)Model=QWEN2-7B, Temperature Scaling=false2026.05 | 38.6 | |
| BaseModel=LLAMA3-8B, Temperature Scaling=false2026.05 | 40.9 | |
| BaseModel=QWEN2-7B, Temperature Scaling=false2026.05 | 44.5 |