Preprint · 2021 · Open access
All Tokens Matter: Token Labeling for Training Better Vision Transformers
arXiv (Cornell University), 34 · Published 22 April 2021 · DOI 10.48550/arxiv.2104.10858
In brief
“All Tokens Matter: Token Labeling for Training Better Vision Transformers” is a 2021 preprint by Zihang Jiang and 7 co-authors, published in arXiv (Cornell University). OpenAlex counts 329 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
- 329
- Most cited in
- 2023 (79)
- References
- 67
- Authors
- 8
Abstract
In this paper, we present token labeling -- a new training objective for training high-performance vision transformers (ViTs). Different from the standard training objective of ViTs that computes the classification loss on an additional trainable class token, our proposed one takes advantage of all the image patch tokens to compute the training loss in a dense manner. Specifically, token labeling reformulates the image classification problem into multiple token-level recognition problems and assigns each patch token with an individual location-specific supervision generated by a machine annotator. Experiments show that token labeling can clearly and consistently improve the performance of various ViT models across a wide spectrum. For a vision transformer with 26M learnable parameters serving as an example, with token labeling, the model can achieve 84.4% Top-1 accuracy on ImageNet. The result can be further increased to 86.4% by slightly scaling the model size up to 150M, delivering the minimal-sized model among previous models (250M+) reaching 86%. We also show that token labeling can clearly improve the generalization capability of the pre-trained models on downstream tasks with dense prediction, such as semantic segmentation. Our code and all the training details will be made publicly available at https://github.com/zihangJiang/TokenLabeling.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Jiang, Z., Hou, Q., Li, Y., Zhou, D., Shi, Y., Jin, X., Wang, A., & Feng, J. (2021). All Tokens Matter: Token Labeling for Training Better Vision Transformers. arXiv (Cornell University), 34. https://doi.org/10.48550/arxiv.2104.10858
MLA 9
Jiang, Zihang, et al. “All Tokens Matter: Token Labeling for Training Better Vision Transformers.” arXiv (Cornell University), vol. 34, 2021. https://doi.org/10.48550/arxiv.2104.10858.
Chicago
Jiang, Zihang, Qibin Hou, Yuan, Li, Daquan Zhou, Yujun Shi, Xiaojie Jin, Anran Wang, and Jiashi Feng. 2021. “All Tokens Matter: Token Labeling for Training Better Vision Transformers.” arXiv (Cornell University) 34. https://doi.org/10.48550/arxiv.2104.10858.
Harvard
Jiang, Z. et al. (2021) ‘All Tokens Matter: Token Labeling for Training Better Vision Transformers’, arXiv (Cornell University), 34. doi:10.48550/arxiv.2104.10858.
IEEE
Z. Jiang et al., “All Tokens Matter: Token Labeling for Training Better Vision Transformers,” arXiv (Cornell University), vol. 34, 2021, doi: 10.48550/arxiv.2104.10858.
BibTeX
@misc{jiang2021tokens,
title = {{All Tokens Matter: Token Labeling for Training Better Vision Transformers}},
author = {Jiang, Zihang and Hou, Qibin and Li, Yuan, and Zhou, Daquan and Shi, Yujun and Jin, Xiaojie and Wang, Anran and Feng, Jiashi},
howpublished = {arXiv (Cornell University)},
year = {2021},
volume = {34},
doi = {10.48550/arxiv.2104.10858},
url = {https://doi.org/10.48550/arxiv.2104.10858}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2021 | 21 | |
| 2022 | 72 | |
| 2023 | 79 | |
| 2024 | 73 | |
| 2025 | 55 | |
| 2026 | 29 |
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- vision transformer
- training objectives
- image classification
- patch tokens
- semantic segmentation
- dense prediction
- model size reduction
- generalization
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?
Jiang, Z., Hou, Q., Li, Y., Zhou, D., Shi, Y., Jin, X., Wang, A., & Feng, J. (2021). All Tokens Matter: Token Labeling for Training Better Vision Transformers. arXiv (Cornell University), 34. https://doi.org/10.48550/arxiv.2104.10858
How many times has this paper been cited?
329 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?
Zihang Jiang, Qibin Hou, Yuan, Li, Daquan Zhou, Yujun Shi, Xiaojie Jin, Anran Wang, Jiashi Feng.
What is the DOI of this paper?
10.48550/arxiv.2104.10858, permanently at https://doi.org/10.48550/arxiv.2104.10858. 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.