| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VirtualHome v2.2.4 (Apartments #3 and #6) | Speedup (%)38.41 | 18 | 1mo ago | ||
| Pro2Bench Average | Qwen3.6-VL-27B | G-Mean F183 | 10 | 1mo ago | |
| Pro2Bench EgoProactive | Qwen3.6-VL-27B | G-Mean F168 | 10 | 1mo ago | |
| Pro2Bench HoloAssist | Llama 4-Maverick | G-Mean F187 | 10 | 1mo ago | |
| Pro2Bench Ego4D | Qwen3.6-VL-27B | G-Mean F186 | 10 | 1mo ago | |
| Pro2Bench HowTo | Qwen3.6-VL-27B | G-Mean F188 | 10 | 1mo ago | |
| Pro2Bench EK | Qwen3.6-VL-27B | G-Mean F10.9 | 10 | 1mo ago | |
| Pro2Bench EE4D | Qwen3.6-VL-27B | G-Mean F189 | 10 | 1mo ago | |
| Household domain | MindZero w/ Qwen3-4B | Speedup19.1 | 10 | 1mo ago | |
| PROACTIVEBENCH (test) | Recall100 | 10 | 5mo ago | ||
| Gridworld | MindZero | Speedup24.5 | 8 | 1mo ago | |
| ProActEval | T100 Score8.11 | 2 | 2mo ago | ||
| ProActEval 200 scenarios | ProactiveAgent | T80 Score5.6 | 2 | 2mo ago |