Maze Navigation on Maze Navigation
86.6Success RateSSL
Evaluation Results
| Method | Links | |
|---|---|---|
| SSLBackbone=Qwen2.5, Model Scale=7B, Reward Strategy=Sweet Spot Learning2026.01 | 86.6 | |
| RL-ContinuousBackbone=Qwen2.5, Model Scale=7B, Reward Strategy=Continuous2026.01 | 85.9 | |
| RL-BinaryBackbone=Qwen2.5, Model Scale=7B, Reward Strategy=Binary2026.01 | 85.7 | |
| SSLBackbone=Qwen2.5, Model Scale=3B, Reward Strategy=Sweet Spot Learning2026.01 | 72.5 | |
| RL-ContinuousBackbone=Qwen2.5, Model Scale=3B, Reward Strategy=Continuous2026.01 | 62.3 | |
| RL-BinaryBackbone=Qwen2.5, Model Scale=3B, Reward Strategy=Binary2026.01 | 55.8 |