Uncertainty Quantification on Longform (PRR)
0.221PRRGLU-S˜∗
Evaluation Results
| Method | Links | |
|---|---|---|
| GLU-S˜∗Model=Fanar2026.06 | 0.221 | |
| GLUModel=Fanar2026.06 | 0.2 | |
| LogProbModel=Fanar2026.06 | 0.083 | |
| LogTokUModel=Fanar2026.06 | 0.077 | |
| RAUQModel=Fanar2026.06 | 0.066 | |
| GLU-EDLModel=Qwen2026.06 | 0.047 | |
| P(true)Model=Gemma2026.06 | 0.004 | |
| P(true)Model=Fanar2026.06 | -0.011 | |
| LogProbModel=Qwen2026.06 | -0.035 | |
| GLUModel=Qwen2026.06 | -0.04 | |
| RAUQModel=Qwen2026.06 | -0.044 | |
| LogTokUModel=Qwen2026.06 | -0.045 | |
| LogProbModel=Gemma2026.06 | -0.189 | |
| GLU-Sα-AUModel=Gemma2026.06 | -0.201 | |
| P(true)Model=Qwen2026.06 | -0.216 | |
| RAUQModel=Gemma2026.06 | -0.234 | |
| GLUModel=Gemma2026.06 | -0.242 | |
| LogTokUModel=Gemma2026.06 | -0.249 |