List your company

Mathematical Reasoning on AIME25 (Average Score)

0.84Average ScoreTool-Star
-0.03360.19320.420.6468Jan 30, 2026

Evaluation results

MethodLinks
Tool-StarCategory=SFT-RL TIR Methods2026.010.84
Tool-Star-SFTCategory=SFT-only TIR Methods2026.010.82
ToRLCategory=RL-only TIR Methods2026.010.79
Vanilla SFT-RL TIRCategory=SFT-RL TIR Methods2026.010.78
AutoTrajCategory=SFT-RL TIR Methods2026.010.76
AutoTIRCategory=RL-only TIR Methods2026.010.67
ReSearchCategory=RL-only TIR Methods2026.010.37
Qwen2.5-7B-InstructFramework=Multi-Dimensional Trajectory Evaluation2026.010
R1-SearcherCategory=RL-only TIR Methods2026.010