Math Tutoring on Math Tutoring
32.5AccuracyNorm-ITPO
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Norm-ITPOAdvantage Estimator=RLOO, Credit Assignment=Norm-ITPO2026.03 | 32.5 | 620 | 29.33 | |
| Norm-ITPOAdvantage Estimator=GRPO, Credit Assignment=Norm-ITPO2026.03 | 31.12 | 840 | 26.83 | |
| LLM as JudgeAdvantage Estimator=GRPO, Credit Assignment=LLM as Judge2026.03 | 30.62 | 1,600 | 22.57 | |
| PRIMEAdvantage Estimator=GRPO, Credit Assignment=PRIME2026.03 | 30.19 | 1,130 | 24.53 | |
| Uniform DecomposeAdvantage Estimator=RLOO, Credit Assignment=Uniform Decompose2026.03 | 30 | 1,180 | 24.01 | |
| PRIMEAdvantage Estimator=RLOO, Credit Assignment=PRIME2026.03 | 29.75 | 1,540 | 21.54 | |
| Trajectory-LevelAdvantage Estimator=RLOO, Credit Assignment=Trajectory-Level2026.03 | 29.06 | 1,440 | 21.82 | |
| ITPOAdvantage Estimator=RLOO, Credit Assignment=ITPO2026.03 | 29.06 | 580 | 26.08 | |
| LLM as JudgeAdvantage Estimator=RLOO, Credit Assignment=LLM as Judge2026.03 | 28.87 | 1,240 | 22.59 | |
| VM+Norm-ITPOAdvantage Estimator=PPO, Credit Assignment=VM+Norm-ITPO2026.03 | 27.15 | 740 | 23.44 | |
| VM OnlyAdvantage Estimator=PPO, Credit Assignment=VM Only2026.03 | 26.81 | 1,360 | 20.06 | |
| Uniform DecomposeAdvantage Estimator=GRPO, Credit Assignment=Uniform Decompose2026.03 | 26.62 | 1,300 | 20.71 | |
| VM+ITPOAdvantage Estimator=PPO, Credit Assignment=VM+ITPO2026.03 | 26.38 | 940 | 21.51 | |
| Trajectory-LevelAdvantage Estimator=GRPO, Credit Assignment=Trajectory-Level2026.03 | 26.37 | 980 | 21.41 | |
| ITPOAdvantage Estimator=GRPO, Credit Assignment=ITPO2026.03 | 26.37 | 580 | 23.51 | |
| VanillaAdvantage Estimator=N/A, Credit Assignment=Vanilla2026.03 | 25.81 | 1,700 | 17.29 | |
| VM+PRIMEAdvantage Estimator=PPO, Credit Assignment=VM+PRIME2026.03 | 22.63 | 1,540 | 14.92 |