# How Human Corrections Travel Across Robots: Safety Issues in Federated VLA Adaptation

- Source: arXiv cs.RO
- Published: 2026-10-26T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/jq06soy5wgdq2hf8n8s7zd8am
- Original: https://arxiv.org/abs/2610.32888

## Summary

Regarding the prospects of combining federated learning with VLA policies, the research points out that sharing human-correction adaptations across tasks, environments, people, and robot embodiments carries safety risks: a shared update that raises average task success may suppress rare safety interventions or produce unsafe responses on another robot.