Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes Paper โข 2608.05000 โข Published 3 days ago โข 51
Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-training Paper โข 2509.26625 โข Published Sep 30, 2025 โข 44
PartGen: Part-level 3D Generation and Reconstruction with Multi-View Diffusion Models Paper โข 2412.18608 โข Published Dec 24, 2024 โข 19