SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features Paper • 2502.14786 • Published Feb 20 • 154
Scaling Text-Rich Image Understanding via Code-Guided Synthetic Multimodal Data Generation Paper • 2502.14846 • Published Feb 20 • 14
RelaCtrl: Relevance-Guided Efficient Control for Diffusion Transformers Paper • 2502.14377 • Published Feb 20 • 12
