AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

Learning Robot Behavior Priors from Non-Robotic Synthetic Data

arXiv · AI, language, vision and robotics · article · Oct 22, 2026 · UTC

Real robot data, rich across tasks and morphologies, is often treated as the bottleneck for training robot learning models, including embodied foundation models that aim for zero-shot physical transfer. In this paper, we investigate the degree to which such models can acquire necessary structure using data from systems without embodiment. To this end, we test if inexpensive, procedurally generated synthetic data can yield useful priors before a model observes any real-world behavior. We study this through robot action prediction, a natural pretraining objective and a direct test of whether str

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-10-23T07:12:42.771Z. This is not the publication date.