UnlockPickup on MiniGrid 4x4 (test)
2,055.08Reward AUCULPS
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| ULPScalibration=uncertainty-aware2026.06 | 2,055.08 | 99.9 | 7.24 | 2.02 | 999 | 7,286 | |
| Linear RL (PPO)Algorithm=PPO2026.06 | 1,865.57 | 74.9 | 15.84 | 1.45 | 749 | 24,412 | |
| Uncalibrated LLMcalibration=none2026.06 | 1,706.43 | 94 | 18.39 | 1.69 | 940 | 20,284 | |
| Q-Learning (GRPO)Algorithm=GRPO2026.06 | 1,515.71 | 82.4 | 16.19 | 1.52 | 824 | 22,142 | |
| DQN (GRPO)Algorithm=GRPO2026.06 | 317.46 | 11.6 | 31.66 | 0.45 | 116 | 47,873 | |
| Unguided RL (PPO)Algorithm=PPO2026.06 | 221.31 | 5.9 | 35.54 | 0.24 | 59 | 49,147 |