Navigation on FrozenLake LLM
30.5Success RateFreshPER
Evaluation results
| Method | Links | |
|---|---|---|
| FreshPERType=LLM2026.04 | 30.5 | |
| On-PolicyType=LLM2026.04 | 29.7 | |
| Standard PERType=LLM2026.04 | 28.1 |
| Method | Links | |
|---|---|---|
| FreshPERType=LLM2026.04 | 30.5 | |
| On-PolicyType=LLM2026.04 | 29.7 | |
| Standard PERType=LLM2026.04 | 28.1 |