VITA-VLA: Efficiently Teaching Vision-Language Models to Act via Action Expert Distillation

Understand

Vision-Language Action (VLA) models significantly advance robotic manipulation by leveraging the strong perception capabilities of pretrained vision-language models (VLMs).

• By integrating action modules into these pretrained models, VLA methods exhibit improved generalization.
• However, training them from scratch is costly.
• In this work, we propose a simple yet effective distillation-based framework that equips VLMs with action-execution capability by transferring knowledge from pretrained small action models.