List your company

Puzzle Solving on Sokoban Simple

2.304ScoreFreshPER
0.420560.909531.39851.88747Apr 18, 2026

Evaluation results

MethodLinks
FreshPERType=LLM2026.042.304
Standard PERType=LLM2026.040.907
On-PolicyType=LLM2026.040.493