Science Reasoning and Simulation on ScienceWorld OOD
0.46Score
Task-Perturbed NLL Optimization
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
| 0.46 | ||
2026.06 | 0.4 | |
| 0.25 | ||
2026.06 | 0.22 |
Task-Perturbed NLL Optimization
| Method | Links | |
|---|---|---|
| 0.46 | ||
2026.06 | 0.4 | |
| 0.25 | ||
2026.06 | 0.22 |