SPoT: Subpixel Placement of Tokens in Vision Transformers
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908869421170688 |
|---|---|
| author | Hjelkrem-Tan, Martine Aasan, Marius Arteaga, Gabriel Y. Rivera, Adín Ramírez |
| author_facet | Hjelkrem-Tan, Martine Aasan, Marius Arteaga, Gabriel Y. Rivera, Adín Ramírez |
| contents | Vision Transformers naturally accommodate sparsity, yet standard tokenization methods confine features to discrete patch grids. This constraint prevents models from fully exploiting sparse regimes, forcing awkward compromises. We propose Subpixel Placement of Tokens (SPoT), a novel tokenization strategy that positions tokens continuously within images, effectively sidestepping grid-based limitations. With our proposed oracle-guided search, we uncover substantial performance gains achievable with ideal subpixel token positioning, drastically reducing the number of tokens necessary for accurate predictions during inference. SPoT provides a new direction for flexible, efficient, and interpretable ViT architectures, redefining sparsity as a strategic advantage rather than an imposed limitation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_01654 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SPoT: Subpixel Placement of Tokens in Vision Transformers Hjelkrem-Tan, Martine Aasan, Marius Arteaga, Gabriel Y. Rivera, Adín Ramírez Computer Vision and Pattern Recognition Machine Learning Vision Transformers naturally accommodate sparsity, yet standard tokenization methods confine features to discrete patch grids. This constraint prevents models from fully exploiting sparse regimes, forcing awkward compromises. We propose Subpixel Placement of Tokens (SPoT), a novel tokenization strategy that positions tokens continuously within images, effectively sidestepping grid-based limitations. With our proposed oracle-guided search, we uncover substantial performance gains achievable with ideal subpixel token positioning, drastically reducing the number of tokens necessary for accurate predictions during inference. SPoT provides a new direction for flexible, efficient, and interpretable ViT architectures, redefining sparsity as a strategic advantage rather than an imposed limitation. |
| title | SPoT: Subpixel Placement of Tokens in Vision Transformers |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2507.01654 |