EXIMO: VLM Guided Exploration of VLA Policies
Ranking
Overall
87
Content
95
Popularity
70
Observed public metrics from 1 member.
Merged summary
TL;DR - EXIMO is a three-stage method for efficiently fine-tuning large vision-language-action robot policies on new manipulation tasks. It combines VLM-guided task decomposition, imitation learning on newly orchestrated data, and residual off-policy reinforcement learning to improve sample efficiency and final performance.
- A vision-language model plans by decomposing long-horizon tasks into shorter subproblems for the VLA policy.
- The planner-policy combination collects an orchestrated dataset, reducing reliance on costly human teleoperation.
- The VLA first learns from this dataset through imitation, then receives further refinement via residual off-policy RL.
- Ablation experiments attribute gains to the explore, imitate, and optimize stages and report significant improvements over existing approaches.
Sources (1)
EXIMO: VLM Guided Exploration of VLA Policies
Public signals
Hugging Face upvotes 16
TL;DR - EXIMO is a three-stage method for efficiently fine-tuning large vision-language-action robot policies on new manipulation tasks. It combines VLM-guided task decomposition, imitation learning on newly orchestrated data, and residual off-policy reinforcement learning to improve sample efficiency and final performance.
- A vision-language model plans by decomposing long-horizon tasks into shorter subproblems for the VLA policy.
- The planner-policy combination collects an orchestrated dataset, reducing reliance on costly human teleoperation.
- The VLA first learns from this dataset through imitation, then receives further refinement via residual off-policy RL.
- Ablation experiments attribute gains to the explore, imitate, and optimize stages and report significant improvements over existing approaches.