Cross-task Generalization on Cooking (test)
0.6889SimilarityOOWM 3-Stage
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| OOWM 3-StageTraining Configuration=3-Stage2026.02 | 0.6889 | 34.47 | 44.48 | 38.24 | |
| Unstructured Baseline2026.02 | 0.6357 | 20.1 | 42.19 | 26.94 | |
| OOWM 2-StageTraining Configuration=2-Stage2026.02 | 0.6058 | 41.19 | 40.59 | 30.76 | |
| Hybrid Strategy2026.02 | 0.5705 | 26.83 | 42.69 | 32.13 |