Play Xylophone on Real-World Tasks Play Xylophone
78Success RateTACO
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TACOIteration=22026.07 | 78 | 97 | |
| TACOIteration=12026.07 | 63 | 115 | |
| TACO (w/o KI)Iteration=12026.07 | 58 | 144 | |
| TACO (w/o KI)Iteration=22026.07 | 52 | 120 | |
| Filtered BCIteration=22026.07 | 51 | 125 | |
| Filtered BCIteration=12026.07 | 49 | 128 | |
| Base PolicyIteration=Base2026.07 | 46 | 132 |