List your company

Mathematical Reasoning on AMC23 (Average Score)

91Average ScoreQwen2.5-7B-Instruct
74.3678.688387.32Jan 30, 2026

Evaluation results

MethodLinks
Qwen2.5-7B-InstructFramework=Multi-Dimensional Trajectory Evaluation2026.0191
R1-SearcherCategory=RL-only TIR Methods2026.0189
AutoTrajCategory=SFT-RL TIR Methods2026.0189
Tool-Star-SFTCategory=SFT-only TIR Methods2026.0188
AutoTIRCategory=RL-only TIR Methods2026.0187
Tool-StarCategory=SFT-RL TIR Methods2026.0182
Vanilla SFT-RL TIRCategory=SFT-RL TIR Methods2026.0180
ReSearchCategory=RL-only TIR Methods2026.0176
ToRLCategory=RL-only TIR Methods2026.0175