# Activation injection study dissects vision-language-action models

- Source: arXiv cs.RO
- Published: 2026-10-21T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/x31g42pymfzkv4rqhclpqq01e
- Original: https://arxiv.org/abs/2603.19233

## Summary

Researchers propose an interpretability pipeline to analyze what a fine-tuned vision-language-action (VLA) policy is computing internally. On pi0.5, injecting task A's activations into the action expert while the policy views task B's scene caused 99.6% of episodes (n=1968) to execute task A's motor trajectory, indicating the injected program is bound to absolute workspace coordinates and ignores the visible scene.

## Why it matters

The study uses activation injection to break down the internal computation of VLA policies, offering a mechanistic explanation for understanding their perturbation brittleness.