List your company

Mathematical Reasoning on AIME 2024 (Pass@1 and Pass@4)

66.87Pass@1AOPD
39.143646.341853.5460.7382May 7, 2026

Evaluation results

MethodLinks
AOPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0566.8779.58
GKDBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0566.2580.65
ExOPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0561.6777.55
OPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0561.4678.61
AOPDBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0561.0475.08
ExOPDBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0559.1773.99
SFT CheckpointBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0558.9674.95
GKDBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0558.1376.61
OPDBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0557.0876.69
GKDBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0554.7977.92
AOPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0553.7577.39
SFT CheckpointBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0552.2971.9
SeqKDBackbone=Qwen3-8B-Base, SFT Warm-up steps=3K2026.0551.0476.56
ExOPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0545.8372.31
SFT CheckpointBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0545.2168.29
OPDBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0543.1367.38
SeqKDBackbone=Qwen3-4B-Base, SFT Warm-up steps=3K2026.0542.565.23
SeqKDBackbone=Qwen3-8B-Base, SFT Warm-up steps=1K2026.0540.2168.08