List your company

General Reasoning on Overall

93.51AccuracyROMCSC
35.73850.736565.73580.7335Mar 23, 2026Apr 2, 2026Apr 13, 2026Apr 23, 2026May 4, 2026May 14, 2026May 25, 2026

Evaluation results

MethodLinks
ROMCSC2026.0393.519471,15915.5712.37
EAT2026.0393.471,5351,8437.576.26
L12026.0393.071,0491,31310.088.1
Cut20482026.0392.771,2821,8427.865.79
ROM2026.0392.531,1881,42111.169.14
Vanilla2026.0391.721,6362,1977.045.6
Cut10242026.0391.299121,74310.155.9
Cut5122026.0384.955131,68016.585.83
MCPO-DAPOBackbone=Qwen3-8B2026.0555.87————
MCPO-GRPOBackbone=Qwen3-8B2026.0554.22————
MCPO-DAPOBackbone=Qwen3-4B2026.0552.6————
MT-GRPOBackbone=Qwen3-8B2026.0552.09————
MGSBackbone=Qwen3-8B2026.0551.77————
MCPO-GRPOBackbone=Qwen3-4B2026.0551.69————
DAPOBackbone=Qwen3-8B2026.0551.68————
GRPOBackbone=Qwen3-8B2026.0550.51————
CLIPOBackbone=Qwen3-8B2026.0550.46————
MT-GRPOBackbone=Qwen3-4B2026.0549.37————
DAPOBackbone=Qwen3-4B2026.0549.24————
MGSBackbone=Qwen3-4B2026.0549.18————
GRPOBackbone=Qwen3-4B2026.0547.95————
CLIPOBackbone=Qwen3-4B2026.0547.22————
Base ModelBackbone=Qwen3-8B2026.0541.34————
Base ModelBackbone=Qwen3-4B2026.0537.96————