Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes Paper • 2608.05000 • Published 7 days ago • 59
CityRAG: Stepping Into a City via Spatially-Grounded Video Generation Paper • 2604.19741 • Published Apr 21 • 18
LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory Paper • 2603.03269 • Published Mar 3 • 63