List your company

Mathematical Reasoning on AMC 2023 (pass@4 accuracy)

75pass@4 AccuracySAGC
30.842.27553.7565.225Jun 1, 2026

Evaluation results

MethodLinks
SAGCBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=Dynamic2026.0675
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=162026.0667.5
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=42026.0667.5
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=82026.0665
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=162026.0662.5
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=82026.0660
SAGCBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=Dynamic2026.0660
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=42026.0655
SAGCBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=Dynamic2026.0647.5
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=162026.0645
SAGCBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=Dynamic2026.0642.5
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=162026.0642.5
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=42026.0640
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=82026.0637.5
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=82026.0635
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=42026.0632.5