List your company

Mathematical Reasoning on MATH 500 (Pass@16, Mean@16)

67.56Mean@16ADS
30.733640.294349.85559.4157Jun 21, 2026

Evaluation results

MethodLinks
ADSBackbone=Olmo3-7B-SFT2026.0667.5688.2
ADSBackbone=Qwen3-4B-Base2026.0667.1384.73
DOTS+RRBackbone=Olmo3-7B-SFT2026.0664.2386.27
DOTS+RRBackbone=Qwen3-4B-Base2026.0663.8584
ADSBackbone=Qwen2.5-Math-1.5B2026.0663.4784.8
DOTS+RRBackbone=Qwen2.5-Math-1.5B2026.0661.6883.33
GRPOBackbone=Qwen3-4B-Base2026.0661.4984.73
GRPOBackbone=Olmo3-7B-SFT2026.0659.3584.27
GRPOBackbone=Qwen2.5-Math-1.5B2026.0655.0482.87
BaseBackbone=Olmo3-7B-SFT2026.0654.7182.67
BaseBackbone=Qwen2.5-Math-1.5B2026.0638.4680.6
BaseBackbone=Qwen3-4B-Base2026.0632.1579.87