List your company

Math Reasoning on AIME25 (Pass@16, Vendi)

31.7Pass@16ESamp
27.5428.6229.730.78Apr 27, 2026

Evaluation results

MethodLinks
ESampBackbone=Qwen2.5-7B-Instruct, parallel sequences=162026.0431.70.46
VanillaBackbone=Qwen2.5-7B-Instruct, parallel sequences=162026.0430.30.32
Min-PBackbone=Qwen2.5-7B-Instruct, parallel sequences=162026.0429.50.3
OverRIDEBackbone=Qwen2.5-7B-Instruct, parallel sequences=162026.0427.70.35