Article

ActiveAlign bridges active perception and VLA policies via observation-aligned data adaptation

The paper proposes ActiveAlign, a data adaptation framework that addresses the mismatch between the initial observation state produced by active camera motion and the manipulation policy's training distribution (first-frame distribution sensitivity): on LIBERO, directly prepending camera repositioning to pi_0.5 reduces the success rate from 46.3% to 11.5%.

Part of the event: Proposing ActiveAlign: A Data Adaptation Framework for Active Perception and VLA Policies

This page provides a summary and source link only.