| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| AGENT new concepts from new initial states (test) | FTL-IGM | Accuracy73 | 9 | 5mo ago | |
| HM3D and MP3D Open-vocabulary objects v0.2 (val) | PLMD† | Success Rate (SR)35.4 | 4 | 2mo ago | |
| MiniGrid RedBall (unseen seeds) | LLM4Teach | Success Rate95.8 | 4 | 5mo ago | |
| MiniGrid RedBlueDoor (unseen seeds) | LLM4Teach | Success Rate95.6 | 4 | 5mo ago | |
| MiniGrid LavaCrossing (unseen seeds) | LLM4Teach | Success Rate93.1 | 4 | 5mo ago | |
| MiniGrid DoorKey (unseen seeds) | LLM4Teach | Success Rate97 | 4 | 5mo ago | |
| Hello Robot Stretch Real-world 3 (test) | Image2Nav | OSR13 | 3 | 17d ago | |
| MoCap human experiment (New Initial State) | FTL-IGM | Percentage Depicts Concept80 | 3 | 5mo ago | |
| MoCap human experiment New Concept | FTL-IGM | Success Rate73.3 | 3 | 5mo ago | |
| MoCap human experiment (train) | FTL-IGM | Success Time Percentage80 | 3 | 5mo ago |