# Learning Foresight without Explicit Trajectories for 3D Diffusion Policies > Source: > Published: 2026-09-21 02:00:01+00:00 3D diffusion policies are strong at generating geometrically grounded actions from current observations, but successful manipulation requires not only knowing what motion is feasible now, but also anticipating where the interaction is heading. Existing policies largely leave such foresight to emerge