Pairwise progress-judgment on RoboPulse Medium hop range
79Accuracy (Real)Robo-Dopamine
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Robo-Dopaminezero-shot=true2026.03 | 79 | 89 | 88 | 84 | 85 | |
| GVLzero-shot=true2026.03 | 71 | 71 | 75 | 69 | 72 | |
| VLACzero-shot=true2026.03 | 67 | 75 | 70 | 75 | 72 | |
| Gemini 3 Pro Previewzero-shot=true2026.03 | 65 | 70 | 73 | 59 | 67 | |
| CLIP ViT-L/14 (I2I)zero-shot=true, backbone=ViT-L/14, input_interface=I2I2026.03 | 61 | 65 | 50 | 59 | 59 | |
| Qwen3-VL-8B-Instructzero-shot=true, parameters=8B, variant=Instruct2026.03 | 61 | 65 | 61 | 41 | 57 | |
| GPT-5.2zero-shot=true2026.03 | 58 | 57 | 54 | 34 | 51 | |
| CLIP ViT-B/32 (I2I)zero-shot=true, backbone=ViT-B/32, input_interface=I2I2026.03 | 56 | 64 | 52 | 50 | 56 | |
| Qwen3-VL-4B-Instructzero-shot=true, parameters=4B, variant=Instruct2026.03 | 56 | 56 | 59 | 50 | 55 | |
| CLIP ViT-B/32 (T2I)zero-shot=true, backbone=ViT-B/32, input_interface=T2I2026.03 | 51 | 48 | 35 | 34 | 42 | |
| CLIP ViT-L/14 (T2I)zero-shot=true, backbone=ViT-L/14, input_interface=T2I2026.03 | 49 | 42 | 35 | 52 | 45 |