VLA^2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept Manipulation Paper • 2510.14902 • Published Oct 16 • 15
