Question Answering on IPhO
40IPhO AccuracySim2Reason
Evaluation Results
| Method | Links | |
|---|---|---|
| Sim2ReasonBase Model=Qwen3-30B2026.04 | 40 | |
| Prime P1 30BBase Model=Qwen3-30B2026.04 | 38.6 | |
| LIMO-32BBase Model=Qwen2.5-32B2026.04 | 25.5 | |
| DAPO-32BBase Model=Qwen2.5-32B2026.04 | 24.7 |