| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| ROBOTHOR | ReAct | Success Rate (SR)40 | 131 | 1mo ago | |
| ALFWorld | EmbodiSkill w/ GPT-5.2 | Success Rate93.28 | 29 | 2mo ago | |
| VLABench Avg. | ERVLA | Progress Score (PS)65.9 | 10 | 1mo ago | |
| VLABench Instruction | ERVLA | Progress Score (PS)70.2 | 10 | 1mo ago | |
| VLABench Commonsense | π0.5 | Progress Score (PS)57.3 | 10 | 1mo ago | |
| VLABench Cross Category | ERVLA | Primitive Success (PS)61 | 10 | 1mo ago | |
| VLABench In-dist. | ERVLA | Progress Score (PS)81.1 | 10 | 1mo ago | |
| VLABench Texture | π0.5 | Progress Score (PS)62.3 | 9 | 1mo ago | |
| MEMENTO Utilization Stage, Joint-memory | Qwen-2.5-7b | Planning Cycles26.9 | 6 | 4mo ago | |
| MEMENTO Utilization Stage, Single-memory | Planning Cycles15.3 | 6 | 4mo ago | ||
| MEMENTO Acquisition Stage | Planning Cycles16 | 6 | 4mo ago |