arxiv:2503.16282
Junlin Han
Junlinh
AI & ML interests
Computer vision, deep learning, artificial intelligence.
Recent Activity
upvoted a paper about 23 hours ago
Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes submitted a paper about 23 hours ago
Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes upvoted a paper about 2 months ago
RepFusion: Leveraging Multimodal Priors for Denoising in Representation SpaceOrganizations
None yet