Uncertainty Quantification on Longform
65.9AUROCGLU-S˜∗
Evaluation Results
| Method | Links | |
|---|---|---|
| GLU-S˜∗Model=Fanar2026.06 | 65.9 | |
| GLUModel=Fanar2026.06 | 65.4 | |
| LogTokUModel=Fanar2026.06 | 56.9 | |
| GLU-EDLModel=Qwen2026.06 | 55.8 | |
| LogProbModel=Fanar2026.06 | 55.8 | |
| RAUQModel=Fanar2026.06 | 54.9 | |
| P(true)Model=Fanar2026.06 | 52.5 | |
| GLUModel=Qwen2026.06 | 49.6 | |
| LogTokUModel=Qwen2026.06 | 49 | |
| LogProbModel=Qwen2026.06 | 48.6 | |
| P(true)Model=Gemma2026.06 | 47.2 | |
| RAUQModel=Qwen2026.06 | 46.9 | |
| GLU-Sα-AUModel=Gemma2026.06 | 39 | |
| P(true)Model=Qwen2026.06 | 38.9 | |
| LogProbModel=Gemma2026.06 | 37.3 | |
| GLUModel=Gemma2026.06 | 35.8 | |
| RAUQModel=Gemma2026.06 | 35.4 | |
| LogTokUModel=Gemma2026.06 | 34.3 |