The DOI matches its Crossref record.
Token Cropr: Faster ViTs for Quite a Few Tasks
Abstract
The adoption of Vision Transformers (ViTs) in resource-constrained applications necessitates improvements in inference throughput. To this end several token pruning and merging approaches have been proposed that improve efficiency by successively reducing the number of tokens. However, it remains an open problem to design a token reduction method that is fast, maintains high performance, and is applicable to various vision tasks. In this work, we present a token pruner that uses auxiliary prediction heads that learn to select tokens end-to-end based on task relevance. These auxiliary heads can be removed after training, leading to throughput close to that of a random pruner. We evaluate our method on image classification, semantic segmentation, object detection, and instance segmentation, and show speedups of 1.5 – 4x with small drops in performance. As a best case, on the ADE20k semantic segmentation benchmark, we observe a 2x speedup relative to the no-pruning baseline, with a negligible performance penalty of 0.1 median mIoU across 5 seeds.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“In vision transformers, it has primarily been used to reduce computational cost by removing visually redundant tokens [12, 13, 31, 49, 73, 76, 85, 111].”
From Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodality (2025) · cites this paper as [12]
Citations
Cites
Still being checked.
Cited by
Sorted by citations
A Hidden Stumbling Block in Generalized Category Discovery: Distracted Attention
Xu, Q., Hu, Z., Duan, Y., Pei, E., & Tai, Y. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00045 · Cited by 6
Fuller, A., Yassin, Y., Wen, J., Ibrahim, T., Kyrollos, D. G., Green, J., & Shelhamer, E. (2025) · neural information processing systems · doi:10.52202/085713-2910 · Cited by 4
Token Compression Meets Compact Vision Transformers: a Survey and Comparative Evaluation for Edge AI
Nguyen, P., & Cheung, N. M. (2025) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.1109/apsipaasc65261.2025.11249396 · Cited by 3
Rényi Attention Entropy for Patch Pruning
Aizawa, H., & Igaue, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31441-3_2 · Cited by 1
Kong, Z., Li, Y., Zeng, F., Xin, L., Messica, S., Lin, X., Zhao, P., Kellis, M., Tang, H., & Zitnik, M. (2025) · arXiv · doi:10.48550/arxiv.2505.18227 · Cited by 51
Metadata
- Authors
- Bergner, B., Lippert, C., & Mahendran, A.
- Affiliations
- Hasso Plattner Institute (DE)
- Google DeepMind (United Kingdom) (GB)
- Venue
- 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2025-06-10
- Pages
- 9740-9750
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4413144648
- Cited by
- 5
- References
- 0
- Citations by year
- 2026: 4 · 2025: 1
- FWCI
- 3.78
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- token pruning, vision transformer, token reduction, inference throughput, resource-constrained applications, semantic segmentation, object detection, instance segmentation, image classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
