AIBullisharXiv – CS AI · 7h ago6/10
🧠
LIBERO-Occ: Evaluating and Improving Vision-Language-Action Models under Scene-Induced Occlusion via Viewpoint Imagination
Researchers introduce LIBERO-Occ, a benchmark for evaluating Vision-Language-Action (VLA) models under object occlusion in robotic manipulation tasks. They propose Viewpoint Imagination (VIM), a technique that generates synthetic alternative viewpoints to improve model robustness when task-relevant objects are partially hidden, achieving performance gains without requiring additional cameras.