Robot Learning with World Models: Capabilities, Frontiers, and Challenges

Researchers compare world models for robot learning, concentrating on physical accuracy, consistency over space and time, and information beyond vision. Talks and contributed work address action models, tactile and proprioceptive signals, evaluation, and planning through imagined interactions. Session: 08:00–17:00 (Australia/Sydney). A paid NeurIPS virtual pass includes workshop livestreams; on-site access follows the host-location registration rules.