List your company

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

39.4Average Score @32SAPO
24.21628.15832.136.042Feb 19, 2026

Evaluation results

MethodLinks
SAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0239.458.1
MASPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0239.358.6
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0238.858.5
BAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0238.357.5
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0237.457.2
DACBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0236.658.7
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-7B2026.0236.554
MASPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0230.754.1
SAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0229.652.7
DACBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0227.454
GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0226.148.8
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0225.248.6
BAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0225.144.7
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0224.849.3