| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Offline Reinforcement Learning | D4RL Gym halfcheetah-medium | Normalized Return74.8 | 44 | |
| Offline Reinforcement Learning | D4RL Gym halfcheetah-medium-expert | Normalized Return114 | 28 | |
| Offline Reinforcement Learning | D4RL Gym halfcheetah-medium-replay | Normalized Average Return69.7 | 27 | |
| Locomotion | D4RL Gym (random-medium) | HalfCheetah Score52.7 | 12 | |
| Locomotion | D4RL Gym medium-expert | HalfCheetah Score96.8 | 12 | |
| Locomotion | D4RL Gym (medium-replay) | HalfCheetah Return53.1 | 12 | |
| Locomotion | D4RL Gym (medium) | HalfCheetah Score51.1 | 12 | |
| Offline Reinforcement Learning | D4RL Gym medium-replay, medium-expert | HalfCheetah (medium-replay)45.7 | 5 |