ProactiveVLA: Augmenting Embodied Memory through Proactive Environment Exploration

What happened

arXiv:2610.06999v1 Announce Type: new Abstract: Rapid adaptation to a new environment requires a robot to acquire (one company buying another) useful knowledge about local objects, states, and interactions from limited experience. After completing an initial task, the agent allocates the remaining interaction budget to self-proposed goals covering object affordances, state-changing interactions, and compositions of interactions.

Systems that combine a reasoning agent with a frozen vision-language-action model (VLA) can adapt through execution feedback and memory, making the choice of experience central to their effectiveness. Repeated practice of a target task may refine a familiar solution while leaving other interactions relevant to changed conditions untested.

It verifies execution outcomes and consolidates both task-directed and exploratory experience into memory that guides subsequent planning and control. ProactiveVLA outperforms the baselines under the same turn budget on LIBERO-Pro and RoboCasa365 Composite-Seen.

Key facts

• arXiv:2610.06999v1 Announce Type: new Abstract: Rapid adaptation to a new environment — requires: a robot to acquire useful knowledge about local objects, states, and interactions from limited experience