Preprint · 2025 · Open access
Differentiable Hierarchical Visual Tokenization
arXiv (Cornell University) · Published 4 November 2025 · DOI 10.48550/arxiv.2511.02652
In brief
“Differentiable Hierarchical Visual Tokenization” is a 2025 preprint by Aasan, Marius and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 2 citations as of 22 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).
- Citations
- 2
- Most cited in
- 2026 (2)
- Authors
- 5
Abstract
Vision Transformers rely on fixed patch tokens that ignore the spatial and semantic structure of images. In this work, we introduce an end-to-end differentiable tokenizer that adapts to image content with pixel-level granularity while remaining backward-compatible with existing architectures for retrofitting pretrained models. Our method uses hierarchical model selection with information criteria to provide competitive performance in both image-level classification and dense-prediction tasks, and even supports out-of-the-box raster-to-vector conversion.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Marius, A., Martine, H.-T., Nico, C., Changkyu, C., & Ramírez, R. A. (2025). Differentiable Hierarchical Visual Tokenization. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2511.02652
MLA 9
Marius, Aasan,, et al. “Differentiable Hierarchical Visual Tokenization.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2511.02652.
Chicago
Marius, Aasan,, Hjelkrem-Tan, Martine, Catalano, Nico, Choi, Changkyu, and Rivera, Adín Ramírez. 2025. “Differentiable Hierarchical Visual Tokenization.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2511.02652.
Harvard
Marius, A. et al. (2025) ‘Differentiable Hierarchical Visual Tokenization’, arXiv (Cornell University). doi:10.48550/arxiv.2511.02652.
IEEE
A. Marius, H.-T. Martine, C. Nico, C. Changkyu, and R. A. Ramírez, “Differentiable Hierarchical Visual Tokenization,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2511.02652.
BibTeX
@misc{marius2025differentiable,
title = {{Differentiable Hierarchical Visual Tokenization}},
author = {Marius, Aasan, and Martine, Hjelkrem-Tan, and Nico, Catalano, and Changkyu, Choi, and Ramírez, Rivera, Adín},
howpublished = {arXiv (Cornell University)},
year = {2025},
doi = {10.48550/arxiv.2511.02652},
url = {https://doi.org/10.48550/arxiv.2511.02652}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- vision transformer
- information criteria
- image classification
- dense prediction
- raster-to-vector conversion
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Marius, A., Martine, H.-T., Nico, C., Changkyu, C., & Ramírez, R. A. (2025). Differentiable Hierarchical Visual Tokenization. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2511.02652
How many times has this paper been cited?
2 times, according to OpenAlex on 22 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Aasan, Marius, Hjelkrem-Tan, Martine, Catalano, Nico, Choi, Changkyu, Rivera, Adín Ramírez.
What is the DOI of this paper?
10.48550/arxiv.2511.02652, permanently at https://doi.org/10.48550/arxiv.2511.02652. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 22 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.