Puzzle Solving on Sokoban Hard
0.847ScoreStandard PER
Evaluation Results
| Method | Links | |
|---|---|---|
| Standard PERType=LLM2026.04 | 0.847 | |
| On-PolicyType=LLM2026.04 | 0.842 | |
| FreshPERType=LLM2026.04 | 0.512 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| Standard PERType=LLM2026.04 | 0.847 | |
| On-PolicyType=LLM2026.04 | 0.842 | |
| FreshPERType=LLM2026.04 | 0.512 |