List your company

Policy Referral on GTAPolicy

9.45GTAPolicy ScoreTriMPI
7.60928.08718.5659.0429Oct 10, 2025

Evaluation results

MethodLinks
TriMPIStrategy=PoRo-GRPO2025.109.45
TriMPIStrategy=PoRo-DAPO2025.109.13
CoT SFT + GRPORL Algorithm=GRPO2025.108.93
CoT SFT + DAPORL Algorithm=DAPO2025.108.73
CoT SFT2025.107.68