Featured

Mechanistic Study of VLA Models: Not All Features Are Created Equal

The study proposes an interpretability pipeline and finds that injecting task A's activations into the action expert causes pi0.5 to execute task A's motor trajectory in 99.6% of episodes (n=1,968), and that the injected program is bound to absolute workspace coordinates while ignoring the visible scene, offering a mechanistic explanation for the perturbation brittleness reported by benchmarks.

Part of the event: Study: pi0.5 actions bind to absolute coordinates

This page provides a summary and source link only.