Uncertainty Quantification on TruthfulQA
68AUROCLogTokU
Evaluation Results
| Method | Links | |
|---|---|---|
| LogTokUModel=Qwen2026.06 | 68 | |
| GLU-EDLModel=Qwen2026.06 | 68 | |
| GLU-AUModel=Fanar2026.06 | 66.1 | |
| RAUQModel=Fanar2026.06 | 64.1 | |
| GLUModel=Fanar2026.06 | 64.1 | |
| LogProbModel=Fanar2026.06 | 63.3 | |
| P(true)Model=Gemma2026.06 | 62.9 | |
| GLUModel=Qwen2026.06 | 61.6 | |
| LogProbModel=Qwen2026.06 | 61.5 | |
| Add-SαModel=Gemma2026.06 | 58.9 | |
| RAUQModel=Qwen2026.06 | 58.2 | |
| LogTokUModel=Fanar2026.06 | 57.9 | |
| P(true)Model=Qwen2026.06 | 55.6 | |
| LogProbModel=Gemma2026.06 | 55.3 | |
| GLUModel=Gemma2026.06 | 54.8 | |
| P(true)Model=Fanar2026.06 | 54.3 | |
| LogTokUModel=Gemma2026.06 | 53.1 | |
| RAUQModel=Gemma2026.06 | 46.6 |