ResearchBenchmarksInverted cart-pole balancingFollow0.0093Violation Rate (Last 100 steps)PPO-L0.0086320.0131410.017650.022159Jun 26, 2026Evaluation ResultsMethodMethodLinksViolation Rate (Last 100 steps)Violation Rate (End Value)Reward (Last 100 steps)Reward (End Value)Success RatePPO-L2026.060.00930.0094597.43597.43—PPO-EAL-m2026.060.00990.0099597.68597.7—PPO-EAL2026.060.010.0099597.22597.15—P3O2026.060.0110.011595.09595.06—PP02026.060.0130.014596.37596.34—PP0-AL2026.060.0260.034596.41596.59—