Safe Reinforcement Learning on Safety Gymnasium PointGoal (C=25) benchmark leaderboard · SOTA2 Research

Safe Reinforcement Learning on Safety Gymnasium PointGoal (C=25)

23.79ReturnCSPO

Evaluation Results

Method	Links
CSPO2026.06		23.79	23.7
IPO2026.06		23.35	27.5
APPO2026.06		23.12	24.3
CPO2026.06		22.58	26.9
P3O2026.06		22.28	24.1
PPO-Lag2026.06		21.78	22
CUP2026.06		20.4	25.3
EPO2026.06		20.19	24.6
PCPO2026.06		19.82	27.1
C-TRPO2026.06		19.34	24.8
FOCOPS2026.06		15.56	30.8
CPPOPID2026.06		6.36	25.1