Workspace Models: Lightweight Robotic Memory via Saliency-Driven Supervision
This paper proposes an alternative approach in which computationally intensive VLM queries are made during train-time to learn a lightweight latent memory that can be efficiently queried at deployment time, and demonstrates that the workspace tokens are not only more lightweight, but also lead to better policy performa...
Nitish Dashora, Douglas Chen, Idan Shenfeld et al.
· 0 citations