Reinforcement Learning on PROCGEN CHASER 1.0 (test)
62.75AccuracyLAOF-Action
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LAOF-ActionAction Supervision=true2025.11 | 62.75 | 0.51 | |
| LAOFAction Supervision=false2025.11 | 53.83 | 0.39 | |
| LAOM-ActionAction Supervision=true2025.11 | 52.95 | 0.43 | |
| CoMo w/ OFOptical Flow (OF) constraints=true2025.11 | 52.53 | 0.29 | |
| CoMo2025.11 | 35.39 | 0.23 | |
| LAPO2025.11 | 26.87 | 0.11 |