| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| TextCraft | Kintsugi | Success Rate (SR)100 | 17 | 1mo ago | |
| AgentGym All (test) | AAWM | Success Rate69.3 | 5 | 1mo ago | |
| AgentGym ALFWorld (test) | AAWM | Success Rate36 | 5 | 1mo ago | |
| AgentGym WebShop (test) | AAWM | Success Rate97.5 | 5 | 1mo ago | |
| AgentGym SciWorld (test) | AAWM | Success Rate91.2 | 5 | 1mo ago | |
| WebShop (OOD) | Task-Perturbed NLL Optimization | Score49.6 | 4 | 1mo ago | |
| ScienceWorld (OOD) | Task-Perturbed NLL Optimization | Score29.6 | 4 | 1mo ago | |
| ALFWorld (OOD) | Task-Perturbed NLL Optimization | Score58.7 | 4 | 1mo ago |