List your company

Graduate-level Professional Science Reasoning on GPQA

32.8Genetics AccuracyHIPPO
21.3624.3327.330.27Jun 28, 2026

Evaluation results

MethodLinks
HIPPOTraining Paradigm=Ours2026.0632.848.8
SP3FTraining Paradigm=Reward-Shaping2026.062845.6
Qwen2.5_7BTraining Paradigm=Standard Training, Configuration=+ SFT2026.0626.644.4
Qwen2.5_7BTraining Paradigm=Standard Training, Configuration=+ RL2026.062544.8
PREF-GRPOTraining Paradigm=Reward-Shaping2026.062541.1
Qwen2.5_7BTraining Paradigm=Standard Training2026.0621.947.9
PRM RLTraining Paradigm=Reward-Shaping2026.0621.848.7