Visual Planning on AVG. (FROZENLAKE, MAZE, MINIBEHAVIOR)
0.806EMVPRL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VPRLInput=Image, Output=Image, post-trained=true, backbone=LVM-7B2025.05 | 0.806 | 0.849 | |
| VPFTInput=Image, Output=Image, post-trained=true, backbone=LVM-7B2025.05 | 0.561 | 0.652 | |
| Qwen 2.5-VL-Instruct-7B (SFT)Input=Text + Image, Output=Text, post-trained=true2025.05 | 0.536 | 0.699 | |
| Gemini 2.5 Pro (think)Input=Text + Image, Output=Text2025.05 | 0.437 | 0.601 | |
| Gemini 2.0 Flash (CoT)Input=Text + Image, Output=Text2025.05 | 0.128 | 0.378 | |
| Gemini 2.0 Flash (Direct)Input=Text + Image, Output=Text2025.05 | 0.101 | 0.363 | |
| Qwen 2.5-VL-Instruct-7B (CoT)Input=Text + Image, Output=Text2025.05 | 0.037 | 0.197 | |
| Qwen 2.5-VL-Instruct-7B (Direct)Input=Text + Image, Output=Text2025.05 | 0.007 | 0.131 |