List your company

Graduate-level Science Reasoning on GPQA (test)

34.3AccuracyE1-Math
26.021628.170830.3232.4692May 16, 2025

Evaluation results

MethodLinks
E1-Mathbackbone=1.5B, budget=(4K,1K)2025.0534.32,758.77
DeepSeek-R1-Distill-Qwenbackbone=1.5B2025.0533.0411,780.87
L1setting=Max(3600)2025.0531.923,892.47
SelfBudgeterbackbone=1.5B2025.0530.653,326.83
E1-Mathbackbone=1.5B, budget=(0.5K,1K)2025.0526.341,278.19