List your company

Mathematical Reasoning on MATH500 (Acc, Len, L-Acc)

88.8AccuracyBingo-A
62.17669.0887682.912Jun 9, 2025

Evaluation results

MethodLinks
Bingo-ABackbone=DeepSeek-R1-Distill-Qwen-7B2025.0688.81,40080.9
PPOBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0688.41,53679.7
Bingo-EBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0687.21,25280.3
BaseBackbone=DeepSeek-R1-Distill-Qwen-7B2025.0682.83,03365.7
Bingo-ABackbone=Qwen2.5-Math-7B-Instruct2025.0682.665673
Bingo-ABackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.0682.289477.6
PPOBackbone=Qwen2.5-Math-7B-Instruct2025.068267072.3
Bingo-EBackbone=Qwen2.5-Math-7B-Instruct2025.0681.655973.6
PPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.0681.42,77166.2
BaseBackbone=Qwen2.5-Math-7B-Instruct2025.0680.872770.3
Bingo-EBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.0680.677976.7
BaseBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.0663.23,91345.7