| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Simulation Benchmark | Hand-Designed Features | Retrieve Object Success Rate (Spatial)68 | 8 | 1mo ago | |
| RoboCasa365 Composite-Unseen | QWEN-ROBOTMANIP | Success Rate14.9 | 7 | 1mo ago | |
| RoboCasa365 Composite-Seen | RLDX-1 | Success Rate27.5 | 7 | 1mo ago | |
| LIBERO-10 Long | UniVLA | Success Rate92 | 7 | 3mo ago | |
| Real-world Robot Workstation Unseen Viewpoints | ICWM | Success Rate (45°)36.6 | 6 | 1mo ago | |
| D4RL FrankaKitchen | AdaptDiff | D4RL Normalized Score58.1 | 6 | 2mo ago | |
| Real-world Long-Horizon Tasks (seen environment) | VLMimic | Task Success Rate: Make coffee40 | 6 | 5mo ago | |
| Real-World Long Horizon Task | SCDP | Open Drawer Success Rate90 | 5 | 1mo ago | |
| BEHAVIOR-1K (Simulation) | ForesightFlow | Radio Score51 | 5 | 1mo ago | |
| RLBench | RoboStream-32B | Bridge Between Towers Success Rate88 | 5 | 4mo ago | |
| RLBench Simulated | SADP | Success Rate (PutRubbish InBin)92.2 | 3 | 2mo ago |