Brick Pose Estimation on BC-Bench Full (test)
65.63Translation Error (LDU)Brick-Composer
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Brick-ComposerBackbone=Qwen-3-8B-VL, Evaluation Protocol=Integrated learning2026.06 | 65.63 | 37.97 | |
| Brick-ComposerBackbone=Gemma-3-12B, Evaluation Protocol=Integrated learning2026.06 | 123.69 | 45.43 | |
| World Feedback (L)Backbone=Qwen-3-8B-VL, Evaluation Protocol=Simulator-generated feedback2026.06 | 137.26 | 52.65 | |
| Designer SupervisionBackbone=Qwen-3-8B-VL, Evaluation Protocol=Human supervision2026.06 | 162.82 | 57.81 | |
| World Feedback (L)Backbone=Gemma-3-12B, Evaluation Protocol=Simulator-generated feedback2026.06 | 170.33 | 51.49 | |
| Designer SupervisionBackbone=Gemma-3-12B, Evaluation Protocol=Human supervision2026.06 | 201.51 | 55.69 | |
| Qwen-3-VL-8BEvaluation Protocol=Zero-shot2026.06 | 210.14 | 62.47 | |
| InternVL-3.5-8BEvaluation Protocol=Zero-shot2026.06 | 221.69 | 88.43 | |
| World Feedback (P)Backbone=Qwen-3-8B-VL, Evaluation Protocol=Simulator feedback2026.06 | 226.33 | 65.66 | |
| Gemma-3-12BEvaluation Protocol=Zero-shot2026.06 | 269.09 | 63 | |
| World Feedback (P)Backbone=Gemma-3-12B, Evaluation Protocol=Simulator feedback2026.06 | 273.41 | 62.26 | |
| GPT-5.4Evaluation Protocol=Zero-shot2026.06 | 310.78 | 74.85 | |
| Qwen-3.5-VL-27BEvaluation Protocol=Zero-shot2026.06 | 314.32 | 82.94 |