Uncertainty Quantification on Multiturn
83.1AUROCGLU
Evaluation Results
| Method | Links | |
|---|---|---|
| GLUModel=Fanar2026.06 | 83.1 | |
| GLU-AUModel=Fanar2026.06 | 83.1 | |
| P(true)Model=Gemma2026.06 | 81 | |
| RAUQModel=Fanar2026.06 | 80.4 | |
| P(true)Model=Qwen2026.06 | 80 | |
| LogProbModel=Fanar2026.06 | 78 | |
| GLU-S¯Model=Qwen2026.06 | 75.8 | |
| GLUModel=Qwen2026.06 | 73.9 | |
| RAUQModel=Gemma2026.06 | 73.4 | |
| LogProbModel=Gemma2026.06 | 73 | |
| GLU-S∗Model=Gemma2026.06 | 71.3 | |
| GLUModel=Gemma2026.06 | 70.1 | |
| P(true)Model=Fanar2026.06 | 69.1 | |
| RAUQModel=Qwen2026.06 | 64.4 | |
| LogProbModel=Qwen2026.06 | 63.5 | |
| LogTokUModel=Qwen2026.06 | 52.3 | |
| LogTokUModel=Fanar2026.06 | 46.3 | |
| LogTokUModel=Gemma2026.06 | 45.7 |