List your company

Mathematical Reasoning on Olympiad-Bench (Avg@32, Pass@32)

58Avg@32MASPO
41.3645.685054.32Feb 19, 2026

Evaluation results

MethodLinks
MASPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.025874.9
DACBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0257.976
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0257.575.5
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0257.273.9
BAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0257.174.5
SAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0256.674.6
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-7B2026.0256.473.7
SAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0248.266.7
MASPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0247.865.7
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0246.464.3
GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0246.364.7
DACBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0246.364.5
BAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.024461.3
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.024261.6