Image tokenizers convert pixels into compact latent sequences for efficient generative modeling. Most approaches arrange latents on a downsampled two-dimensional grid, tying sequence length to image resolution. We compare two variable-length strategies: computationally efficient patch-derived tokens, which inherit explicit spatial organization, and more demanding register tokens, which have no predefined spatial layout. Our reconstruction results reveal that patch-derived tokens perform substantially better under extreme compression, whereas register tokens achieve slightly better reconstruction quality at intermediate budgets. At high token budgets, both approaches achieve similar reconstruction quality. For downstream generative tasks, register tokens outperform patch-derived tokens. Adding representation alignment with external pretrained visual encoders substantially improves the generative performance of patch-derived tokens, allowing them to match or outperform register tokens depending on the token budget. We further find that both approaches learn spatially localized representations. Patch-derived tokens largely retain their explicitly imposed spatial organization, while register tokens develop an implicit two-dimensional spatial structure despite having no predefined layout. Representation alignment reduces this spatial organizational constraint, enabling individual tokens to exert a stronger non-local influence.
UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation
Shanghai Jiao Tong University(上海交通大学);Shanghai AI Laboratory(上海人工智能实验室);Beihang University(北京航空航天大学);Shenzhen Key Lab of Computer Vision and Pattern Recognition, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(深圳计算机视觉与模式识别重点实验室,深圳先进技术研究院,中国科学院);Nanjing University(南京大学);University of Science and Technology of China(中国科学技术大学);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院)
Huazhong University of Science and Technology(华中科技大学);Horizon Robotics;Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系)
Peng Cheng Laboratory, Shenzhen, China(鹏城实验室,深圳,中国);Department of Electrical Engineering, City University of Hong Kong, Hong Kong SAR(香港城市大学电子工程系,香港特别行政区);Department of Computer Science, City University of Hong Kong, Hong Kong SAR(香港城市大学计算机科学系,香港特别行政区);School of Computer Science, Peking University, Beijing, China(北京大学计算机学院,北京,中国)
Visual-Word Tokenizer: Beyond Fixed Sets of Tokens in Vision Transformers
Predictive Analytics Lab, University of Sussex, UK(苏塞克斯大学预测分析实验室);University of Surrey, UK(塞夫尔大学);Basque Center for Applied Mathematics, Spain(巴斯克应用数学中心);Monash University, Indonesia(墨尔本大学)
An Adversarial Approach to Register Extreme Resolution Tissue Cleared 3D Brain Images
School of Computer Science, Queensland University of Technology(昆士兰科技大学计算机科学学院);School of Electrical Engineering and Robotics, Queensland University of Technology(昆士兰科技大学电气工程与机器人学院);Centre for Data Science, Queensland University of Technology(昆士兰科技大学数据科学中心)
ARC Lab(ARC实验室);Peking University(北京大学);Huazhong University of Science and Technology(华中科技大学);Foundation Technology Center(基础技术中心);Tsinghua University(清华大学);Tencent PCG(腾讯平台与内容事业群)
Parameter-Efficient CLIP Adaptation for 3D Understanding via Unified Tokenization
Fondazione Bruno Kessler(布鲁诺·科塞拉基金会);University of Trento(特伦托大学);University of Pisa(比萨大学);Beijing Forestry University(北京林业大学);Shanghai Jiao Tong University(上海交通大学);Hong Kong University of Science and Technology (GZ)(香港科技大学);Shandong University(山东大学);University of California, Merced(加州大学默塞德分校)
Deep Learning for Generating Computational PIN-4 Immunohistochemistry Staining from Prostate Biopsy H&E Images
Biomedical Engineering University of California, Irvine Irvine, CA, USA(生物医学工程 卡罗来纳大学伊文城分校 伊文城,加州,美国);Laboratory Medicine Biomedical Engineering Electrical Engineering(实验室医学 生物医学工程 电气工程);Computer Science University of California, Irvine Irvine, CA, USA(计算机科学 卡罗来纳大学伊文城分校 伊文城,加州,美国)
CaTok: Taming Mean Flows for One-Dimensional Causal Image Tokenization
Institute of Trustworthy Embodied AI, Fudan University(可信具身人工智能研究所,复旦大学);Shanghai Innovation Institute(上海创新研究院);Shanghai Key Laboratory of Multimodal Embodied AI(上海多模态具身人工智能重点实验室)
Comparison of Patch-Based Conditional Generative Adversarial Neural Net Models with Emphasis on Model Robustness for Use in Head and Neck Cases for MR-Only planning
Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing
IBM Research(IBM研究院);The University of Alabama in Huntsville(阿拉巴马大学亨茨维尔分校);University Space Research Association (USRA)(大学空间研究协会);NASA Goddard Space Flight Center(美国国家航空航天局戈达德太空飞行中心);SETI Institute(搜寻地外文明研究所);NASA Ames Research Center(美国国家航空航天局艾姆斯研究中心);University of Maryland, Baltimore County (UMBC)(马里兰大学巴尔的摩县分校);Howard University(霍华德大学)