List your company

Question Answering on SimpleQA-verified OOD

42.2AccuracyVanilla Search-R1
20.3626.0331.737.37Jan 12, 2026

Evaluation results

MethodLinks
Vanilla Search-R1Model=Qwen3-4B2026.0142.20.2870.210.874
Temperature ScalingModel=Qwen3-4B2026.0142.20.2540.1980.874
CAR (MSCR)Model=Qwen3-4B2026.0141.80.1060.1060.929
MASHModel=Qwen2.5-7B2026.0141.50.4650.420.775
CAR (MSCR)Model=Qwen2.5-7B2026.0140.90.150.1560.837
Vanilla Search-R1Model=Qwen2.5-7B2026.0140.70.4410.3980.776
Temperature ScalingModel=Qwen2.5-7B2026.0140.70.3850.3470.776
CAR (Weighted Brier, λ=1/3)Model=Qwen2.5-7B2026.0140.50.1770.1760.798
CAR (Weighted Brier, λ=1/3)Model=Qwen3-4B2026.0140.10.1270.1270.912
MASHModel=Qwen3-4B2026.0137.40.2990.2130.859
CAR (MSCR)Model=Qwen2.5-3B2026.0136.60.1920.1910.763
CAR (Weighted Brier, λ=1/3)Model=Qwen2.5-3B2026.0135.80.2030.2030.74
MASHModel=Qwen2.5-3B2026.0135.10.5890.560.696
Vanilla Search-R1Model=Qwen2.5-3B2026.0134.80.610.5870.702
Temperature ScalingModel=Qwen2.5-3B2026.0134.80.5830.5490.702
CAR (Weighted Brier, λ=1)Model=Qwen3-4B2026.0133.10.0360.0330.973
CAR (Weighted Brier, λ=1)Model=Qwen2.5-7B2026.0124.40.050.0530.94
CAR (Weighted Brier, λ=1)Model=Qwen2.5-3B2026.0121.20.0270.0270.983