Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces EVIS, an Experience-Guided and Behavior-Validated Initiation Search framework to efficiently discover reliable initiation configurations for deploying frozen learned skills in new environments. EVIS leverages historical execution data to prune unpromising candidates and employs target-environment behavior validation to confirm their efficacy, mitigating the high cost of extensive real-world rollouts. Evaluation on manipulation tasks demonstrates EVIS's ability to significantly reduce interaction queries and improve reliable initiation discovery under limited budgets.
Frozen skills can be reliably deployed in new environments with 80% fewer target-environment queries by intelligently combining historical experience with targeted behavioral validation.
Deploying frozen learned skills, such as Vision-Language-Action (VLA) policies, in new environments requires identifying initiation configurations that support reliable execution. In skill composition, an initiation configuration affects not only the current skill but also the physical state passed to subsequent skills, so successful execution of an individual skill does not necessarily imply successful completion of the composed task. Estimating target-specific capability through extensive rollouts is costly in real-world deployment, while directly reusing historical experience can be unreliable under environment changes. We propose EVIS, an Experience-Guided and Behavior-Validated Initiation Search framework for discovering reliable initiation configurations under limited target interaction. EVIS uses historical execution experience to prioritize promising candidates and target-environment behavior to validate whether they remain effective. We evaluate EVIS on single-skill and two-stage manipulation tasks with frozen VLA policies. EVIS reduces mean target-environment queries and improves reliable candidate discovery under small interaction budgets. These results show that combining historical guidance with target-side behavioral validation can reduce the interaction cost of deploying frozen learned skills in new environments.