List your company

Mathematical Reasoning on GSM8K (MV, EM, PK)

75.04Mean Value (MV)I-PPO
42.259250.769659.2867.7904Apr 2, 2026

Evaluation results

MethodLinks
I-PPOBackbone=Gemma2026.0475.0468.4185.23
I-PPOBackbone=LLaMa2026.0474.8368.2690.35
PPOBackbone=Gemma2026.0472.9463.9884.39
PPOBackbone=LLaMa2026.0470.8656.1488.39
GRPOBackbone=Gemma2026.0470.7362.583.02
BaselineBackbone=Gemma2026.0470.1361.9583.02
GRPOBackbone=LLaMa2026.0468.2145.1888.25
BaselineBackbone=LLaMa2026.0465.5834.9988.25
I-PPOBackbone=Rho2026.0451.9346.174.83
PPOBackbone=Rho2026.0450.0541.1173.09
GRPOBackbone=Rho2026.0445.1440.8372.51
BaselineBackbone=Rho2026.0443.5240.2672.48