| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Causal-Plan-Bench in-domain | Causal Planner | Overall Success Rate45.28 | 16 | 1mo ago | |
| EB-Habitat (unseen) | Avg Success Rate57 | 15 | 1mo ago | ||
| ALFWorld Seen v1 (in-distribution) | ProCeedSFT | Average Success Rate57.14 | 14 | 3mo ago | |
| ALFWorld Text Aligned (N=134 unseen tasks) | EvoPlan | Success Rate98.5 | 11 | 17d ago | |
| ALFRED | LoTA (Full Recompute) | Success Rate (SR)45.81 | 11 | 4mo ago | |
| AlfWorld (out of distribution) | ProCeedSFT | Accuracy58.95 | 8 | 3mo ago | |
| AI2-THOR | SVLL-Stage 3 | SR78.35 | 8 | 4mo ago | |
| ALFWorld Text N=134 unseen tasks Misaligned | Token Count501,049 | 7 | 17d ago | ||
| MRoom-30k 1.0 (test) | OOWM 3-Stage | Similarity56.94 | 6 | 3mo ago | |
| ALFWorld OOD v1 | GFlowVLM w/ SubTB | Success Rate12.3 | 6 | 5mo ago | |
| WorldLines 21 action-dense samples downstream planning probe | ObsMem | Plan Judge68.4 | 5 | 1mo ago | |
| ScanNet 3D-LLM | LSceneLLM | ROUGE47.05 | 4 | 5mo ago | |
| 9 real-world robotic tasks zero-shot | SVLL-Stage 3 | Success Rate (SR)55.56 | 2 | 4mo ago |