SOURCE-LINKED INTELLIGENCE

DREAM: Deployment-Time Demonstration Generation via Real-to-Sim for Scalable Policy Adaptation

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Vision-language-action (VLA) models have made strong progress in language-conditioned robot manipulation, but improving their performance in a new workspace still often requires action-labeled data from that environment. Collecting such data by human teleoperation is costly, especially when each workspace, object arrangement, or task may require new demonstrations. We present DREAM, a framework that generates fine-tuning data for a pretrained VLA from a captured workspace and a language instruction, without requiring a task-specific human demonstration. DREAM reconstructs the workspace, automa

Read original source ↗ Open in workspace

recordType

paper

region

Global