List your company

Mathematical Reasoning on CollegeMath (MV, EM, PK)

52.56Mean Value (MV)I-PPO
15.608825.201934.79544.3881Apr 2, 2026

Evaluation results

MethodLinks
I-PPOBackbone=Qwen2.5 3B2026.0452.5643.9973.5
I-PPOBackbone=Phi3 4B2026.0449.0643.8265.44
I-PPOBackbone=Gemma2 2B2026.0448.3836.4462.62
I-PPOBackbone=LLaMa3 8B2026.0448.3144.1566.17
PPOBackbone=Qwen2.5 3B2026.0448.1834.469.94
PPOBackbone=Phi3 4B2026.0445.734.163.84
PPOBackbone=Gemma2 2B2026.0444.7828.3360.54
PPOBackbone=LLaMa3 8B2026.0443.5232.2260.93
BaselineBackbone=Qwen2.5 3B2026.0437.2226.3163.13
BaselineBackbone=Phi3 4B2026.0435.5929.2259.01
BaselineBackbone=Gemma2 2B2026.0434.1720.2955.07
I-PPOBackbone=Rho 1B2026.0431.2528.1747.89
BaselineBackbone=LLaMa3 8B2026.0430.7720.5956.81
PPOBackbone=Rho 1B2026.0428.4621.8545.29
BaselineBackbone=Rho 1B2026.0417.0310.4629.66