Multi-Task Reinforcement Learning (LTL Instruction Following) on ZoneEnv Infinite Horizon
633.33Average VisitsStructLTL
Evaluation results
| Method | Links | |
|---|---|---|
| StructLTLformula=psi_3, zero-shot=true2026.02 | 633.33 | |
| StructLTLformula=psi_6, zero-shot=true2026.02 | 624.44 | |
| StructLTLformula=psi_5, zero-shot=true2026.02 | 606.67 | |
| DeepLTLformula=psi_3, zero-shot=true2026.02 | 560.58 | |
| StructLTLformula=psi_4, zero-shot=true2026.02 | 511.48 | |
| DeepLTLformula=psi_6, zero-shot=true2026.02 | 475.01 | |
| DeepLTLformula=psi_5, zero-shot=true2026.02 | 467.83 | |
| DeepLTLformula=psi_4, zero-shot=true2026.02 | 404.11 | |
| StructLTLformula=psi_1, zero-shot=true2026.02 | 6.72 | |
| DeepLTLformula=psi_1, zero-shot=true2026.02 | 4.77 | |
| StructLTLformula=psi_2, zero-shot=true2026.02 | 2.83 | |
| DeepLTLformula=psi_2, zero-shot=true2026.02 | 1.5 |