Multi-step planning and manipulation on Multi-step grocery bagging
95.24Success RateDIRECT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DIRECTPlanner=DIRECT2026.06 | 95.24 | 6.85 | |
| Qwen3.5-VL 9B TPlanner=Qwen3.5-VL 9B T2026.06 | 90.48 | 19.58 | |
| Qwen3.5-VL 9B IPlanner=Qwen3.5-VL 9B I2026.06 | 47.62 | 2.19 |