Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction Paper • 2610.12299 • Published 8 days ago • 68
GRACE: Generation-aware latent compression for efficient video generation Paper • 2610.10524 • Published 9 days ago • 88
Tetris3D: 3D Scene Generation With Objects That Fit Together Paper • 2610.10539 • Published 9 days ago • 48
CtrlVTON: Controllable Virtual Try-On via Visual-Instance-Prompt Segmentation Paper • 2607.09362 • Published Jul 10 • 8
DA-Flow: Degradation-Aware Optical Flow Estimation with Diffusion Models Paper • 2603.23499 • Published Mar 24 • 52