WorldToken: Time-First Sequence Modeling for Robotic Imitation Learning

Chunkai Yang · hf · 2026-08-25

WorldToken introduces a time-first sequence modeling approach for robotic imitation learning. It fuses heterogeneous robot observations into per-timestep world tokens processed by a causal Transformer and diffusion action head. The paper includes scaling and temporal-context analyses on RoboCasa and RMBench benchmarks.

Original post →

More from Embodied

Embodied channel →