Preprint
Aug 2026
ForeTime-VLA: Causal Future-Token Distillation from a World Action Model for Conveyor-Belt Manipulation
The agreement between offline orientation gains and reduced real-robot contact-pose failures supports causal future-token distillation as an effective way to improve dynamic manipulation without deploying the world-model teacher.
Si-Yuan Ma, Yu-Tian Zhang, Bo-Shi Zhang et al.
· 2 citations