World-model research identifies limits in learning hidden physical parameters
A controlled study found prediction objectives determine whether robot world models retain mass, stiffness, drag, and force information.
Semiconductor Domain: sensor fusion
RoboCup coverage highlights widening tasks beyond robot football
RoboCup 2026 showcased humanoids and service robots addressing household, logistics, and human-assistance tasks beyond competition football.
Spatial Intelligence Is Decomposed for Physical AI Evaluation
Researchers propose more granular evaluation of spatial reasoning needed for physical manipulation and navigation.
World Action Planner links prediction directly to decisions
The World Action Planner uses action-conditioned world models to improve generalizable robot decision-making.
