List your company

Math Reasoning on AIME 2025 (Avg@12)

72.2Average ScorePAINT
34.86444.55754.2563.943Apr 29, 2026

Evaluation results

MethodLinks
PAINTBackbone=Qwen3-8B Instruct2026.0472.2
OPSDBackbone=Qwen3-8B Instruct2026.0470.8
PAINTBackbone=Qwen3-4B Instruct2026.0469.7
GRPOBackbone=Qwen3-8B Instruct2026.0468.9
OPSDBackbone=Qwen3-4B Instruct2026.0468.3
GRPOBackbone=Qwen3-4B Instruct2026.0468.1
No updateBackbone=Qwen3-4B Instruct2026.0466.4
No updateBackbone=Qwen3-8B Instruct2026.0465.6
SFTBackbone=Qwen3-8B Instruct2026.0464.2
SFTBackbone=Qwen3-4B Instruct2026.0462.3
PAINTBackbone=Qwen3-1.7B Instruct2026.0444.2
OPSDBackbone=Qwen3-1.7B Instruct2026.0443.9
GRPOBackbone=Qwen3-1.7B Instruct2026.0438.3
No updateBackbone=Qwen3-1.7B Instruct2026.0436.7
SFTBackbone=Qwen3-1.7B Instruct2026.0436.3