Instruction Following on ALFWorld Hard
20.66Success Rate (SR)TCOD (B2F)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TCOD (B2F)Student Model=Qwen2.5-7B2026.04 | 20.66 | 27.07 | |
| TCOD (F2B)Student Model=Qwen2.5-7B2026.04 | 18.18 | 27.37 | |
| TCOD (B2F)Student Model=Qwen2.5-3B2026.04 | 13.22 | 28.16 | |
| Vanilla OPDStudent Model=Qwen2.5-7B2026.04 | 13.22 | 27.89 | |
| Vanilla OPDStudent Model=Qwen2.5-3B2026.04 | 10.74 | 28.64 | |
| TCOD (F2B)Student Model=Qwen2.5-3B2026.04 | 9.92 | 28.57 | |
| SFTStudent Model=Qwen2.5-7B2026.04 | 8.26 | 28.73 | |
| Qwen2.5-7B-RL (Teacher)Model Type=Teacher2026.04 | 6.61 | 27.31 | |
| SFTStudent Model=Qwen2.5-3B2026.04 | 4.96 | 29.12 | |
| Zero-ShotStudent Model=Qwen2.5-7B2026.04 | 1.65 | 29.77 | |
| Zero-ShotStudent Model=Qwen2.5-3B2026.04 | 0.83 | 29.88 |