List your company

Mathematical Reasoning on SVAMP (Accuracy, Micro avg.)

90.33Accuracy (SVAMP)TraceLift
32.0947.2162.3377.45May 5, 2026

Evaluation results

MethodLinks
TraceLiftModel=Qwen2.5-7B2026.0590.3370.19
Exec-onlyModel=Qwen2.5-14B2026.0588.6771.93
TraceLiftModel=Qwen2.5-14B2026.0588.6772.25
BaseModel=Qwen2.5-14B2026.058671.2
Exec-onlyModel=Qwen2.5-7B2026.0567.6765.62
BaseModel=Qwen2.5-7B2026.0558.6746.51
TraceLiftModel=Llama3.1-8B2026.0547.6722.86
Exec-onlyModel=Llama3.1-8B2026.0546.3322.19
BaseModel=Llama3.1-8B2026.0534.3312.74