List your company

Question Answering on Distractor hard

62.2Accuracy (Distractor hard)NWCAD
28.50437.2524654.748Apr 17, 2026

Evaluation results

MethodLinks
NWCADModel=Llama-3.1-70B2026.0462.2
BaselineModel=Llama-3.1-70B2026.0458.4
With-contextModel=Llama-3.1-70B2026.0457.8
NWCADModel=Ministral-3-8B2026.0455.83
AdaCADModel=Llama-3.1-70B2026.0454.8
AdaCADModel=Ministral-3-8B2026.0453.21
With-contextModel=Ministral-3-8B2026.0452.82
NWCADModel=Llama-3.1-8B2026.0452.16
CoCoAModel=Llama-3.1-70B2026.0452
With-contextModel=Llama-3.1-8B2026.0451.25
AdaCADModel=Llama-3.1-8B2026.0448.49
BaselineModel=Ministral-3-8B2026.0441.42
CADModel=Ministral-3-8B2026.0441.02
BaselineModel=Llama-3.1-8B2026.0440.81
CoCoAModel=Ministral-3-8B2026.0440.5
CoCoAModel=Llama-3.1-8B2026.0439.71
CADModel=Llama-3.1-8B2026.0432.5
CADModel=Llama-3.1-70B2026.0429.8