PredVLA: Predictive Sensorimotor Modeling for Sub-Million-Parameter Robot Manipulation
Research
Source: arXiv cs.ROPublish time unverified
arXiv:2608.26673v2 Announce Type: replace Abstract: Large pretrained vision-language-action models achieve strong robot-manipulation performance, while compact alternatives have largely pursued efficiency by compressing the prevailing observation-to-action paradigm. We investigate whether predictive sensorimotor modeling can make more effective use of a limited parameter budget than direct observation-to-action mapping.