List your company

Question Answering on O.O.D. Averaged (6 datasets)

56.2AccuracyRLCR
53.18453.96754.7555.533Jul 22, 2025

Evaluation results

MethodLinks
RLCR2025.0756.2680.210.21
RLVR2025.0753.9500.460.46
RLVR + BCE Classifier2025.0753.9580.270.24
RLVR + Brier Classifier2025.0753.9600.320.33
RLVR + Probe2025.0753.9530.380.38
Answer Probability2025.0753.9600.420.42
Base2025.0753.3540.410.4