Truthfulness on TruthfulQA (Sel-AUC, AUROC)
0.998Sel-AUCStructU
Evaluation results
| Method | Links | ||
|---|---|---|---|
| StructUModel=Claude 4.5 Sonnet, Aggregation=total2026.06 | 0.998 | 0.953 | |
| Self-ConsUModel=DeepSeek R12026.06 | 0.94 | 0.844 |
| Method | Links | ||
|---|---|---|---|
| StructUModel=Claude 4.5 Sonnet, Aggregation=total2026.06 | 0.998 | 0.953 | |
| Self-ConsUModel=DeepSeek R12026.06 | 0.94 | 0.844 |