List your company

Complex Reasoning on SCoRE (test)

16.26AccuracySelfBudgeter
6.30728.891111.47514.0589May 16, 2025

Evaluation results

MethodLinks
SelfBudgeterbackbone=1.5B2025.0516.264,491.35
L1setting=Max(3600)2025.0513.695,145.91
E1-Mathbackbone=1.5B, budget=(4K,1K)2025.0512.233,327.37
DeepSeek-R1-Distill-Qwenbackbone=1.5B2025.0510.1411,695.94
E1-Mathbackbone=1.5B, budget=(0.5K,1K)2025.056.691,272.04