Robotic Manipulation on LIBERO-Long (test)
100Task 0 Success Rateπ0.5 + TACO
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| π0.5 + TACOSampling Strategy=TACO, Result Source=Official results from TACO [Yang et al., 2025a]2026.05 | 100 | 96 | 98 | 100 | 98 | 100 | 92 | 100 | 86 | 96 | 96.6 | |
| π0.5Result Source=Official results from TACO [Yang et al., 2025a]2026.05 | 98 | 100 | 98 | 98 | 98 | 100 | 96 | 94 | 68 | 98 | 94.8 | |
| π0.5Note=Paper baseline reproduction2026.05 | 98 | 98 | 100 | 100 | 96 | 96 | 94 | 100 | 92 | 94 | 96.8 | |
| π0.5 + TapSamplingSampling Strategy=TapSampling (Ours)2026.05 | 98 | 96 | 100 | 100 | 100 | 100 | 98 | 100 | 90 | 98 | 98 |