Loading paper
VQ-Transplant: Efficient VQ-Module Integration for Pre-trained Visual Tokenizers — Minlit