Science Simulation on ScienceWorld
15.44Success RateBase
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| BaseStudent Model=None, Teacher Model=Qwen3-32B2026.06 | 15.44 | -17.92 | 18.67 | |
| BaseStudent Model=None, Teacher Model=Qwen3-8B2026.06 | 10.74 | -16.09 | 21.67 | |
| SAGE-OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B2026.06 | 7.38 | -16.01 | 20.91 | |
| SAGE-OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-8B2026.06 | 6.71 | 0.24 | 25.3 | |
| Entropy-Aware OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-32B2026.06 | 6.04 | -26.1 | 20 | |
| SAGE-OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-32B2026.06 | 6.04 | -26.1 | 20 | |
| TCOD-F2BStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B2026.06 | 6.04 | -14.89 | 22.02 | |
| OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-8B2026.06 | 4.7 | -2.78 | 25.12 | |
| TCOD-F2BStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-8B2026.06 | 4.7 | -9.68 | 24.06 | |
| OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B2026.06 | 4.03 | -29.89 | 19.64 | |
| BaseStudent Model=Qwen3-0.6B, Teacher Model=None2026.06 | 3.36 | -43.51 | 21.38 | |
| BaseStudent Model=Qwen3-1.7B, Teacher Model=None2026.06 | 3.36 | -24 | 21.38 | |
| Entropy-Aware OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-8B2026.06 | 3.36 | -11.52 | 23.44 | |
| Entropy-Aware OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-8B2026.06 | 3.36 | -3.13 | 25.91 | |
| OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-32B2026.06 | 3.36 | -28.46 | 20.79 | |
| Entropy-Aware OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B2026.06 | 3.36 | -14.99 | 23.59 | |
| TCOD-F2BStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-8B2026.06 | 3.03 | -2.07 | 24.88 | |
| TCOD-F2BStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-32B2026.06 | 2.68 | -33.42 | 23.2 | |
| SFTStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-8B2026.06 | 2.01 | -31.84 | 23.94 | |
| OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-8B2026.06 | 2.01 | -12.88 | 23.94 | |
| SAGE-OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-8B2026.06 | 2.01 | 3.89 | 26.94 | |
| SFTStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-8B2026.06 | 2.01 | -17.56 | 23.28 | |
| SFTStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-32B2026.06 | 2.01 | -34.59 | 18.1 | |
| SFTStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B2026.06 | 2.01 | -15.61 | 23.23 |