List your company

Math Reasoning on AIME 2024 (Avg@12)

79.2Avg@12 ScorePAINT
47.16855.48463.872.116Apr 29, 2026

Evaluation results

MethodLinks
PAINTBackbone=Qwen3-8B Instruct2026.0479.2
OPSDBackbone=Qwen3-8B Instruct2026.0477.8
PAINTBackbone=Qwen3-4B Instruct2026.0476.9
GRPOBackbone=Qwen3-8B Instruct2026.0476.4
OPSDBackbone=Qwen3-4B Instruct2026.0476.4
No updateBackbone=Qwen3-8B Instruct2026.0475.8
GRPOBackbone=Qwen3-4B Instruct2026.0475.6
No updateBackbone=Qwen3-4B Instruct2026.0474.9
SFTBackbone=Qwen3-8B Instruct2026.0472.3
SFTBackbone=Qwen3-4B Instruct2026.0470.2
PAINTBackbone=Qwen3-1.7B Instruct2026.0459.7
OPSDBackbone=Qwen3-1.7B Instruct2026.0457.2
No updateBackbone=Qwen3-1.7B Instruct2026.0451.5
GRPOBackbone=Qwen3-1.7B Instruct2026.0451.1
SFTBackbone=Qwen3-1.7B Instruct2026.0448.4