| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Safety-Gym SafetyPointCircle1 (evaluation) | CUP | Average Reward44.3 | 15 | 25d ago | |
| Safety-Gym SafetyPointGoal1 (evaluation) | PPO | Average Reward11.37 | 11 | 2mo ago | |
| Extended Chain CMDP (last 1,000 episodes) | Unconstrained | Jc2 Constraint Metric0.069 | 3 | 3mo ago | |
| Grid-world Time-Variant Safety Threshold (100 randomly generated environments) | MASE | Safety Violations0 | 2 | 5mo ago | |
| Grid-world Time-Invariant Safety Threshold (100 randomly generated environments) | SNO-MDP | Safety Violation Count0 | 2 | 5mo ago |