Hallucination Detection on Self held-out 50% (test)
61AUROCLlama-3-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| Llama-3-8BEnsemble Strategy=Best J + Best BB2026.06 | 61 | |
| Llama-3-8BEnsemble Strategy=Best J2026.06 | 59 | |
| Llama-3-8BEnsemble Strategy=Best J + Best WB2026.06 | 59 | |
| Llama-3-8BEnsemble Strategy=Weighted Avg.2026.06 | 54.3 | |
| GPT-4oEnsemble Strategy=Best J2026.06 | 51.1 | |
| GPT-4oEnsemble Strategy=Best J + Best BB2026.06 | 50.9 | |
| Gem.-1.5-FlEnsemble Strategy=Best J2026.06 | 50.1 | |
| Gem.-1.5-FlEnsemble Strategy=Best J + Best BB2026.06 | 49.9 | |
| GPT-4oEnsemble Strategy=Best J + Best WB2026.06 | 47.3 | |
| Gem.-1.5-FlEnsemble Strategy=Best J + Best WB2026.06 | 45.9 | |
| GPT-4oEnsemble Strategy=Weighted Avg.2026.06 | 45.1 | |
| Gem.-1.5-FlEnsemble Strategy=Weighted Avg.2026.06 | 40.8 |