SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem Paper • 2609.07064 • Published 6 days ago • 91
RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting Paper • 2609.07414 • Published 6 days ago • 12
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs Paper • 2609.03820 • Published 10 days ago • 16