CLiFT: Compressive Light-Field Tokens for Compute-Efficient and Adaptive Neural Rendering Paper • 2507.08776 • Published Jul 11 • 54
Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning Paper • 2506.13654 • Published Jun 16 • 44
PhysTwin: Physics-Informed Reconstruction and Simulation of Deformable Objects from Videos Paper • 2503.17973 • Published Mar 23 • 8