List your company

Question Answering on Hotpots

88.2AccuracyNWCAD
24.2440.84557.4574.055Apr 17, 2026

Evaluation results

MethodLinks
NWCADModel=Ministral-3-8B2026.0488.2
AdaCADModel=Ministral-3-8B2026.0487.8
With-contextModel=Ministral-3-8B2026.0487.6
With-contextModel=Llama-3.1-70B2026.0487.38
AdaCADModel=Llama-3.1-70B2026.0486.88
NWCADModel=Llama-3.1-8B2026.0486.7
With-contextModel=Llama-3.1-8B2026.0486.4
NWCADModel=Llama-3.1-70B2026.0486.25
AdaCADModel=Llama-3.1-8B2026.0485.7
CADModel=Ministral-3-8B2026.0485.1
CoCoAModel=Llama-3.1-70B2026.0484.75
CoCoAModel=Llama-3.1-8B2026.0483.6
CADModel=Llama-3.1-8B2026.0480
CADModel=Llama-3.1-70B2026.0474.5
CoCoAModel=Ministral-3-8B2026.0474.5
BaselineModel=Llama-3.1-8B2026.0429.1
BaselineModel=Llama-3.1-70B2026.0426.75
BaselineModel=Ministral-3-8B2026.0426.7