Better Source, Better Flow: Learning Condition-Dependent Source Distribution for Flow Matching Paper • 2602.05951 • Published Feb 5
CAMEO: Correspondence-Attention Alignment for Multi-View Diffusion Models Paper • 2512.03045 • Published Dec 2, 2025 • 3
i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models Paper • 2606.11289 • Published 28 days ago • 17
Scenes as Objects, Not Primitives: Instance-Structured 3D Tokenization from Unposed Views Paper • 2606.29513 • Published 9 days ago • 50
MVTrack4Gen: Multi-View Point Tracking as Geometric Supervision for 4D Video Generation Paper • 2606.26087 • Published 13 days ago • 35
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning Paper • 2606.13673 • Published 26 days ago • 110
Geometry-Aware Representation Denoising for Robust Multi-view 3D Reconstruction Paper • 2605.26230 • Published May 25 • 41
WorldKV: Efficient World Memory with World Retrieval and Compression Paper • 2605.22718 • Published May 21 • 42
TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking Paper • 2605.12587 • Published May 12 • 37
DA-Flow: Degradation-Aware Optical Flow Estimation with Diffusion Models Paper • 2603.23499 • Published Mar 24 • 51
Repurposing Geometric Foundation Models for Multi-view Diffusion Paper • 2603.22275 • Published Mar 23 • 48
WorldCam: Interactive Autoregressive 3D Gaming Worlds with Camera Pose as a Unifying Geometric Representation Paper • 2603.16871 • Published Mar 17 • 61
Grounding World Simulation Models in a Real-World Metropolis Paper • 2603.15583 • Published Mar 16 • 155