Fetching the paper…

Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models · Around