Question Answering on PopQA (CF, EM, F1)
94.27Correct First (CF)FIDES
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| FIDESModel=LLaMA3-70B2026.06 | 94.27 | 54.12 | 63.45 | |
| FIDESModel=LLaMA2-13B2026.06 | 90.35 | 42.92 | 44.86 | |
| AdaCADModel=LLaMA3-70B2026.06 | 89.15 | 42.95 | 49.32 | |
| CADModel=LLaMA3-70B2026.06 | 86.54 | 31.02 | 35.12 | |
| AdaCADModel=LLaMA2-13B2026.06 | 78.43 | 35.26 | 39.92 | |
| Standard RAGModel=LLaMA3-70B2026.06 | 75.82 | 39.43 | 43.06 | |
| CADModel=LLaMA2-13B2026.06 | 70.15 | 28.12 | 30.54 | |
| Standard RAGModel=LLaMA2-13B2026.06 | 63.85 | 31.92 | 35.46 |