HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models Paper • 2512.09928 • Published Dec 10, 2025 • 14