List your company

Logical Reasoning on KK All

53.6AccuracyOracle Verifier
3.6816.6429.642.56May 31, 2026

Evaluation results

MethodLinks
Oracle VerifierModel Size=12B2026.0553.6
RevisionSEModel Size=12B2026.0552.8
SimpleSEModel Size=12B, Verifier threshold (τ)=0.62026.0551.1
SimpleSEModel Size=12B, Verifier threshold (τ)=0.72026.0550.1
SimpleSEModel Size=12B, Verifier threshold (τ)=0.52026.0548
gemma-3-12b-itModel Size=12B2026.0547.5
SimpleSEModel Size=12B, Verifier threshold (τ)=0.82026.0547.5
Oracle VerifierModel Size=4B2026.0546.6
RevisionSEModel Size=4B2026.0542.2
SimpleSEModel Size=4B, Verifier threshold (τ)=0.62026.0540.7
SimpleSEModel Size=4B, Verifier threshold (τ)=0.82026.0539.7
SimpleSEModel Size=4B, Verifier threshold (τ)=0.72026.0539.6
SimpleSEModel Size=4B, Verifier threshold (τ)=0.52026.0538.4
gemma-3-4b-itModel Size=4B2026.0531
Oracle VerifierModel Size=1B2026.0512.5
SimpleSEModel Size=1B, Verifier threshold (τ)=0.82026.058.4
gemma-3-1b-itModel Size=1B2026.057.8
RevisionSEModel Size=1B2026.057.8
SimpleSEModel Size=1B, Verifier threshold (τ)=0.72026.056.5
SimpleSEModel Size=1B, Verifier threshold (τ)=0.52026.055.7
SimpleSEModel Size=1B, Verifier threshold (τ)=0.62026.055.6