1 paper · 1 filter
Zhen Liu, Xinyu Ning, Zhe Hu +8
Recent vision-language-action (VLA) systems have demonstrated strong capabilities in embodied manipulation. However, most existing VLA policies rely on limited observation windows…