4 papers
Vision Foundation Models as Effective Visual Tokenizers for Autoregressive Image Generation
Anlin Zheng, Xin Wen, Xuanyang Zhang +5
In this work, we present a novel direction to build an image tokenizer directly on top of a frozen vision foundation model, which is a largely underexplored area. Specifically, we…
"Principal Components" Enable A New Language of Images
Xin Wen, Bingchen Zhao, Ismail Elezi +2
We introduce a novel visual tokenization framework that embeds a provable PCA-like structure into the latent token space. While existing visual tokenizers primarily optimize for re…
A Data-Centric Revisit of Pre-Trained Vision Models for Robot Learning
Xin Wen, Bingchen Zhao, Yilun Chen +2
Pre-trained vision models (PVMs) are fundamental to modern robotics, yet their optimal configuration remains unclear. Through systematic evaluation, we find that while DINO and iBO…
What Makes CLIP More Robust to Long-Tailed Pre-Training Data? A Controlled Study for Transferable Insights
Xin Wen, Bingchen Zhao, Yilun Chen +2
Severe data imbalance naturally exists among web-scale vision-language datasets. Despite this, we find CLIP pre-trained thereupon exhibits notable robustness to the data imbalance…