List your company

Mathematical Reasoning on AMC23 (Avg@32, Pass@32)

91.4Average Score @32MASPO
76.52880.38984.2588.111Feb 19, 2026

Evaluation results

MethodLinks
MASPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0291.495
BAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0289.297.2
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-7B2026.028996.6
SAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0288.795
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0288.196.6
DACBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0288.199.6
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-7B2026.0287.796.6
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0282.894.9
SAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0282.794.8
MASPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0282.295
DACBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0280.295
BAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0280.194.9
GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0279.594.8
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0277.194.1