Junfei Xiao

Latest

• What If We Recaption Billions of Web Images with LLaMA-3?
• Rejuvenating image-GPT as Strong Visual Representation Learners
• Masked Autoencoders Enable Efficient Knowledge Distillers