Question Answering on NQ SYNTH
79AccuracyNWCAD
Evaluation Results
| Method | Links | |
|---|---|---|
| NWCADModel=Llama-3.1-70B2026.04 | 79 | |
| AdaCADModel=Llama-3.1-70B2026.04 | 76.38 | |
| AdaCADModel=Ministral-3-8B2026.04 | 72.8 | |
| NWCADModel=Ministral-3-8B2026.04 | 71.8 | |
| CoCoAModel=Llama-3.1-70B2026.04 | 71.62 | |
| With-contextModel=Ministral-3-8B2026.04 | 70.7 | |
| NWCADModel=Llama-3.1-8B2026.04 | 70 | |
| AdaCADModel=Llama-3.1-8B2026.04 | 69.8 | |
| With-contextModel=Llama-3.1-8B2026.04 | 69.6 | |
| With-contextModel=Llama-3.1-70B2026.04 | 69.5 | |
| CADModel=Llama-3.1-70B2026.04 | 66.62 | |
| CADModel=Ministral-3-8B2026.04 | 64.8 | |
| CoCoAModel=Llama-3.1-8B2026.04 | 64.4 | |
| CADModel=Llama-3.1-8B2026.04 | 64.3 | |
| CoCoAModel=Ministral-3-8B2026.04 | 58.9 | |
| BaselineModel=Llama-3.1-8B2026.04 | 25.5 | |
| BaselineModel=Ministral-3-8B2026.04 | 20.5 | |
| BaselineModel=Llama-3.1-70B2026.04 | 19.25 |