Learned predictive models of environment dynamics.
Luca Zanatta, Grzegorz Malczyk, Kostas Alexis
World models, learned generative models that predict how an environment evolves, have become a promising tool for sample-efficient robot learning. Yet how robust they are to environmental variability remains poorly understood.
Andy Tang, William Chen, Andrew Wagenmaker et al.
Generalist policies can learn a wide range of skills from diverse robot datasets. In order to solve or improve on challenging news tasks, we need a way to infer and invoke the appropriate actions from the policy's rich behavioral prior, especially when directly commanding the policy fails.
Liao, Qiayuan, Shao, Yiyang, Long, Junfeng et al.
Tracking large human-motion datasets has become the dominant route to general, expressive humanoid whole-body control, but scaling across many motions is expensive.
Read these papers in order to build expertise in World Models.
Luca Zanatta, Grzegorz Malczyk et al. · 2026
Andy Tang, William Chen et al. · 2026
Liao, Qiayuan, Shao, Yiyang et al. · 2026
