Safety-constrained Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
44.3Average Reward
15
Jul 1, 2026
11.37Average Reward
11
May 15, 2026
0.069Jc2 Constraint Metric
3
Apr 7, 2026
0Safety Violations
2
Feb 26, 2026
0Safety Violation Count
2
Feb 26, 2026