Math Tutoring on MTBench MathTutorBench OOD
8.29Score (Sc)MIPROv2
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MIPROv2enables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Gen.2026.05 | 8.29 | 8.03 | 8.16 | |
| GEPAenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Gen.2026.05 | 8.23 | 8 | 8.12 | |
| ACEenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Ped.2026.05 | 8.21 | 8.06 | 8.13 | |
| EvoPromptenables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Gen.2026.05 | 8.16 | 8.09 | 8.13 | |
| TextGradenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Ped.2026.05 | 8.01 | 7.72 | 7.86 | |
| TF-GRPOenables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Gen.2026.05 | 7.95 | 7.7 | 7.82 | |
| Frameenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Ped.2026.05 | 7.89 | 7.72 | 7.81 | |
| MetaBlendenables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Gen.2026.05 | 7.88 | 7.63 | 7.76 | |
| OPROenables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Ped.2026.05 | 7.84 | 7.65 | 7.75 | |
| CondBridgeenables thinking (Th.)=true, applies thinking reward (Th.R)=true, prompt seed type=Gen.2026.05 | 7.82 | 7.84 | 7.83 | |
| ParetoGradenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Gen.2026.05 | 7.77 | 7.79 | 7.78 | |
| LeakShieldenables thinking (Th.)=true, applies thinking reward (Th.R)=false, prompt seed type=Gen.2026.05 | 7.37 | 7.87 | 7.62 | |
| No optimizationenables thinking (Th.)=false, applies thinking reward (Th.R)=false, prompt seed type=Gen.2026.05 | 7.17 | 5.25 | 6.21 |