π_{0.5}: a Vision-Language-Action Model with Open-World Generalization Paper • 2504.16054 • Published Apr 22, 2025 • 5