List your company

Mathematical Reasoning on AIME 25 (Avg@32, Pass@32)

42.9Avg@32MASPO
23.1428.2733.438.53Feb 19, 2026

Evaluation results

MethodLinks
MASPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0242.90.732
BAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0237.70.584
GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0237.40.605
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0237.40.672
DACBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0236.20.617
SAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0235.30.58
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-7B2026.0234.40.581
Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0230.10.558
DACBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0228.70.512
MASPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0228.40.58
BAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02280.557
SAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02280.499
GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0227.70.499
Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0223.90.442