X-Planner: event-structured task planning for embodied intelligence

A 32-author VLA planning front-end that makes intermediate planning structure explicit: one VLM backbone exposes both an interpretable discrete interface emitting event states and a latent interface passing continuous chain-of-thought states through staggered Transformer depths ('Staircase Decoding'), anchored by a frozen latent-to-text reconstruction objective. Trained on ego/UMI/teleoperation data with takeover-time annotations, and it beat baselines in real-robot experiments.