AHA: A Vision-Language-Model for Detecting and Reasoning Over Failures in Robotic Manipulation

aha-vlm.github.io

关键词：Foundation Model, Affordance Prediction